History of Napkin Folding

History of Napkin Napkins have been in used from the times of ancient Roman Empire and prior to them, in ancient Greece. References to word napkin  dates back to 1384 AD. Spartans from ancient Greece (around 4th century BC)are known to have used bread slices made from a lump of dough as napkins to wipe their hands and it is possible that they ate these bread napkins after use. The ancient Romans (first to 5th century AD) are known to have used cloth napkins (called Mappa) to protect from food spill and wipe mouth. The guests brought their own napkins and carried away left-over delicacies in their Mappa.With the fall of Roman Empire, napkins disappeared from the dining table. Napkins returned to adorn the dining table many centuries later and the classic painting  Last Supper  from 1464-1467 AD by Dieric Bouts depicts the use of Napkins on the dinner table. By 16th century, napkins were part of rich dining experience and came in many sizes, known by various names like diaper, se rviette, touaille (for towel) depending on the size and intended use. 17th century saw the use of big sized napkins measuring 35inches by 45 inches to help accommodate the needs of eating with bare hands instead of spoon/forks.The size reduced when forks and spoons were accepted as part of regular dining experience in major parts of Europe in the 18th century including Great Britain. The 17th century also saw French come up with elaborate rules for nobility class which included instructions on napkin usage, a predecessor to modern day napkin etiquette, including the one which instructs the guests to not use the napkin for wiping the face or clean teeth or worse, rub nose. Asians – Japanese, Thai, Chinese and Indians have not been using napkins during the meal.While Indians are known to use bare hands for meal and rely on water to wash their hands before and after each meal, Chinese and Japanese use those magical chopsticks which seem to pick up everything they eat, keeping th eir hands clean and stain free. Types of Napkins Napkins are made of paper or cloth that is made of cotton or linen. Clothes are quite popular in formal dining while fast foods and certain restaurants too tend to serve paper napkins. Big occasions like weddings often go for paper napkins since sheer number of guests attending the wedding makes cloth napkins less practical.In general, cloth napkins are usually the choice for those who care for ambience and the environment. To me, cloth napkins are environment friendly while exuding a feeling of luxury. Paper Napkin with Monograms It is often possible to have paper napkins ordered for marriages (and other special meal occasions) with names or initials of the bride and bridegroom embossed on the napkin. Customers are given the option of having custom colors, custom patterns, custom monograms imprinted on the napkin. The size of napkin provided at table varies from occasion and place of dining.Smaller size napkins of about 5 inches x 5 inches are used when serving drinks or starters/appetizers or snacks. For lunch, larger sized napkins measuring about 14 inches to 20 inches square size are provided. Napkins supplied for dinner are even bigger, often of size 20 inches to 22 inches square. The idea is to increase the size of the napkin as the quantity of food/drinks served/consumed increases. This helps take care of the increased risk of spillage with increasing quantity of food and drinks. As a host, when providing napkins to your guests or customers, you need a achieve a good balance of size and utility of the napkins.Napkins which are too big in size for the purpose can only increase laundry load for little or no soil/spills on the napkins. Napkins are made of cotton or linen or linen-cotton mix, although it is possible that a few napkins are made of other materials like synthetic or silk. They come in variety of colors although white seems to be the most popular choice of color. I prefer white napkins over other s since it is easy to detect stains early with white napkins. Restaurant set with White Napkins I often see white napkins being used in restaurants.White conveys cleanliness since it is easy for guests/customers to see the care you have taken to ensure the napkin is clean, fresh and washed. Often napkins and table cloths are sold together as a set to form one matching set of clothes adorning the dining table. Napkin Folding Aside from being conscious about color, size and texture of the napkin used for dining, discerning hosts who care for ambience often go for various  napkin folds  that help the guests feel very welcome, convey a feeling that you value their company, that you are willing to go the extra mile to make the occasion a happy one.A great napkin fold adds to the setting. Some of the folds can be so creative and nice that it is difficult for us to make up our mind to unfold them during the start of the meal. This site has a number of napkin folds which will provide fo r a warm, friendly environment at lunch/dinner time. A few are easy yet elegant while others could require a little more time and skill. Pyramid Napkin Fold Some are more formal while others are pretty casual ones. Some of the folds take just a few minutes of your time and even those who have never folded before can do them relatively quickly.Start with the easy ones and then move on to more difficult folds. Some of these folds require stiff clothing or starching while ironing or washing. Try out multiple folds to see which ones you like or prefer. Apart from the many popular napkin folds listed here, you may find it useful to visit a resource dedicated to paper folding Origami for more ideas on how to fold napkins. If you plan to try out complicated Origami folds, be mindful of the fact that an elegant napkin fold on a lunch/dinner table is less about demonstrating your mind skills than it is about setting the mood for a hearty meal.In other words, do not overdo napkin folding. His tory * Creative napkin folding may have started during the reign of Louis XIV, when people decided to present napkins as an art form. However, napkin folding really took off around the turn of the last century. Function * A creatively folded napkin can make the simplest meal classier, and dinner guests are always impressed by an artfully folded napkin resting at their table setting. * Straight Vinyl SkirtingVinyl with velcro hook designed to drape over the foam edge to floor. www. tiffinmats. com Significance * The art of napery folding has been compared to origami, and many different types of three-dimensional figures may be produced. Types * You can learn to fold napkins into pyramids, pockets for flatware, flowers, hats, fans, slippers and various animals. Considerations * The best type of cloth napkin to use for napkin folding is crisp linen, as it holds its shape better than other types of cloth

Small Towns Big City

Costumes Bring out Different Personalities Pulling up in the parking lot there are small groups of people crowded around getting out of their cars and preparing to enter the building, there is laughter and talking throughout the parking lot leading up to the building. Everyone is dressed in costumes for the theme of the party; there are mummies, zombies, cartoon characters, fairies, devils, musicians, super Heroes, cowboys and cowgirls, and many more. Everyone wearing a costume seems to present a more outgoing and somewhat â€Å"false† personality relevant to the character or thing that their costume is portraying.As we walk closer to the entrance from the parking lot you can hear the music, and the noise of the people inside. The music is loud and draws a crowd towards the doors. When we arrive at the entrance to the building there is a lady about five feet five inches tall with light brown shoulder length hair wearing a cowgirl outfit, standing behind a booth, checking Ident ification to ensure everyone who enters is at least twenty one years of age, and also collecting the entrance fee from each person. She’s talking with a fake country accent that helps add character to her costume.Her personality is bubbly and welcoming; it sets the tone for what the atmosphere is inside the building. There is another person at the entrance, a man standing beside her he is tall about six foot five or so, he has short brown hair and is wearing a uniform shirt instead of a costume. His personality is more calm and natural; I feel that this is because he is not in a costume so he doesn’t have the same ramp up as the others who are in costume have. He checks everyone’s bags for weapons, alcohol, and anything that would not be permitted inside the building.He then places a paper bracelet around each person wrist to identify that they have been checked through security, and are permitted to enter the building, this also helps the bartenders identify th at the person they might be serving alcohol to have been verified as over the age of twenty one. When we walk in the front door the noise of laughing, talking and the music has gotten much louder, it is now clear what the atmosphere in the building is, there are people everywhere, talking, drinking, and having a good time, it seems fun and relax, entertainment is everywhere.No matter what direction you turn there are so many different smells, but with the huge crowed of people the strongest of all the smells would by far be the body odor. Straight ahead as soon as you walk in the door there is a bar where a bar tendered dressed in a devil costume stands taking orders and making drinks, she seems busy and more mute then the rest of the employees, but every once in awhile she will scowl in a fun way at someone in character to her costume.There is a large line of people surrounding the bar waiting for their turn they are all laughing and some are even in a playful mood using props to t heir costumes as weapons to battle with each other. To the left is a long table of people, a dance floor, and a DJ station. Looking to the right there is more small grouped tables, restrooms, and a door to an outside area where people are smoking and talking. There is a group of people standing outside dressed as the ghost buster crew, they are all acting out scenes from the movie, they are sure to stay in full character through the entire scene.Their enthusiasm is entertaining to say the least, and it draws a crowd of people. Others start to join their lead, there is a small group of pirates that are also sward fighting and talking in accents that suggest their characters. The outside atmosphere is more playful then inside and is also a little quieter. Overall everyone is having a good time and the theme of the party is in honor of the holiday Halloween.There is costume contest that everyone was encouraged to enter, it seems to ramp up the excitement throughout the building, everyo ne begins to gather around the stage. They had everyone who entered the costume contest line up on stage one by one. The costumes were judged by the crowed in a scale of how loud their yelling was. As each person was presented in the line of contestants they acted in a way to present the character of their costumes. There was a woman dressed up as a lion tamer, and another woman with her dressed as a lion.They acted out a scene when it was their turn on stage. The tamer attempted to tame the wild lion while the lion persistently tried to attack the tamer. As the night goes on the atmosphere changes, everyone has obviously had a lot more to drink and the levels of intelligence have diminished. As I look to my left by the dance floor I see a fight between two girls, there are shoes flying, and a crowd of people quickly gathering to get in on the action. This is now becoming an atmosphere that I may not want to be in for much longer.Although the festivities are quiet entertaining and a drenaline is pumping with every move from the dance floor it is now clear that the night is almost over. As the bouncers approach the crowed they move through aggressively to break up the fight, and sort out what has happened. I watch intently to see the damages that are done its almost sad how entertaining this has become. Once the fight has been broken up, those involved in the fight, and the bouncers all leave out of the building, the party then goes on.The DJ makes an announcement that this will be the last song and the party will soon end. As the last song comes to an end the crowd of people being to leave out the front door. Everyone is still playful and maintain their characters as they gather around their vehicles and begin to reverse the entire process. Overall the night was entertaining to say the least the costumes created an atmosphere that was out of the normal and would most likely not have existed if it wasn’t for the costume and the theme of the party. Clear a nd engaging introduction that clearly states a strong thesis: 14/20 While engaging and full of detail, there is no clear thesis or map. Something like, â€Å"this was a great party because of the atmosphere, the costumes, and the mood† would have worked. * Clearly established supporting details, in the form of observations, that support your thesis: 15/20 There are lots of supporting details, but because the framework is missing, it is unclear what those details support. * Transitions and Topic Sentences: 15/20Don’t forget topic sentences at the start of every major point. They need to provide a transition, a connection to the thesis, and the topic/point of the paragraph/section. Therefore, when you start discussing atmosphere something like â€Å"Another reason why the party was so much fun was the atmosphere† would have worked. * A conclusion that reinforces the focus and dominant ideas in the text: 18/20 * Engaging Tone and Style:5/5 * Grammar, Punctuation, a nd Spelling: 8/10 * MLA Header and title: 5/5 GRADE: 80

Bayesian Inference

Biostatistics (2010), 11, 3, pp. 397–412 doi:10. 1093/biostatistics/kxp053 Advance Access publication on December 4, 2009 Bayesian inference for generalized linear mixed models YOUYI FONG Downloaded from http://biostatistics. oxfordjournals. org/ at Cornell University Library on April 20, 2013 Department of Biostatistics, University of Washington, Seattle, WA 98112, USA ? HAVARD RUE Department of Mathematical Sciences, The Norwegian University for Science and Technology, N-7491 Trondheim, Norway JON WAKEFIELD? Departments of Statistics and Biostatistics, University of Washington, Seattle, WA 98112, USA [email  protected] ashington. edu S UMMARY Generalized linear mixed models (GLMMs) continue to grow in popularity due to their ability to directly acknowledge multiple levels of dependency and model different data types. For small sample sizes especially, likelihood-based inference can be unreliable with variance components being particularly difficult to estimate. A Bayesian approach is appealing but has been hampered by the lack of a fast implementation, and the difficulty in specifying prior distributions with variance components again being particularly problematic.Here, we briefly review previous approaches to computation in Bayesian implementations of GLMMs and illustrate in detail, the use of integrated nested Laplace approximations in this context. We consider a number of examples, carefully specifying prior distributions on meaningful quantities in each case. The examples cover a wide range of data types including those requiring smoothing over time and a relatively complicated spline model for which we examine our prior specification in terms of the implied degrees of freedom.We conclude that Bayesian inference is now practically feasible for GLMMs and provides an attractive alternative to likelihood-based approaches such as penalized quasi-likelihood. As with likelihood-based approaches, great care is required in the analysis of clustered bina ry data since approximation strategies may be less accurate for such data. Keywords: Integrated nested Laplace approximations; Longitudinal data; Penalized quasi-likelihood; Prior specification; Spline models. 1.I NTRODUCTION Generalized linear mixed models (GLMMs) combine a generalized linear model with normal random effects on the linear predictor scale, to give a rich family of models that have been used in a wide variety of applications (see, e. g. Diggle and others, 2002; Verbeke and Molenberghs, 2000, 2005; McCulloch and others, 2008). This flexibility comes at a price, however, in terms of analytical tractability, which has a ? To whom correspondence should be addressed. c The Author 2009. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals. [email  protected] rg. 398 Y. F ONG AND OTHERS number of implications including computational complexity, and an unknown degree to which inference is dependent on modeling assumptions. Lik elihood-based inference may be carried out relatively easily within many software platforms (except perhaps for binary responses), but inference is dependent on asymptotic sampling distributions of estimators, with few guidelines available as to when such theory will produce accurate inference. A Bayesian approach is attractive, but requires the specification of prior distributions which is not straightforward, in particular for variance components.Computation is also an issue since the usual implementation is via Markov chain Monte Carlo (MCMC), which carries a large computational overhead. The seminal article of Breslow and Clayton (1993) helped to popularize GLMMs and placed an emphasis on likelihood-based inference via penalized quasi-likelihood (PQL). It is the aim of this article to describe, through a series of examples (including all of those considered in Breslow and Clayton, 1993), how Bayesian inference may be performed with computation via a fast implementation and with guidance on prior specification. The structure of this article is as follows.In Section 2, we define notation for the GLMM, and in Section 3, we describe the integrated nested Laplace approximation (INLA) that has recently been proposed as a computationally convenient alternative to MCMC. Section 4 gives a number of prescriptions for prior specification. Three examples are considered in Section 5 (with additional examples being reported in the supplementary material available at Biostatistics online, along with a simulation study that reports the performance of INLA in the binary response situation). We conclude the paper with a discussion in Section 6. 2.T HE G ENERALIZED LINEAR MIXED MODEL GLMMs extend the generalized linear model, as proposed by Nelder and Wedderburn (1972) and comprehensively described in McCullagh and Nelder (1989), by adding normally distributed random effects on the linear predictor scale. Suppose Yi j is of exponential family form: Yi j |? i j , ? 1 ? p(â₠¬ ¢), where p(†¢) is a member of the exponential family, that is, p(yi j |? i j , ? 1 ) = exp yi j ? i j ? b(? i j ) + c(yi j , ? 1 ) , a(? 1 ) Downloaded from http://biostatistics. oxfordjournals. org/ at Cornell University Library on April 20, 2013 for i = 1, . . . , m units (clusters) and j = 1, . . , n i , measurements per unit and where ? i j is the (scalar) ? canonical parameter. Let ? i j = E[Yi j |? , b i , ? 1 ] = b (? i j ) with g(? i j ) = ? i j = x i j ? + z i j b i , where g(†¢) is a monotonic â€Å"link† function, x i j is 1 ? p, and z i j is 1 ? q, with ? a p ? 1 vector of fixed ? Q effects and b i a q ? 1 vector of random effects, hence ? i j = ? i j (? , b i ). Assume b i |Q ? N (0, Q ? 1 ), where ? the precision matrix Q = Q (? 2 ) depends on parameters ? 2 . For some choices of model, the matrix Q is singular; examples include random walk models (as considered in Section 5. ) and intrinsic conditional ? autoregressive models. We further assume tha t ? is assigned a normal prior distribution. Let ? = (? , b ) denote the G ? 1 vector of parameters assigned Gaussian priors. We also require priors for ? 1 (if not a constant) and for ? 2 . Let ? = (? 1 , ? 2 ) be the variance components for which non-Gaussian priors are ? assigned, with V = dim(? ). 3. I NTEGRATED NESTED L APLACE APPROXIMATION Before the MCMC revolution, there were few examples of the applications of Bayesian GLMMs since, outside of the linear mixed model, the models are analytically intractable.Kass and Steffey (1989) describe the use of Laplace approximations in Bayesian hierarchical models, while Skene and Wakefield Bayesian GLMMs 399 (1990) used numerical integration in the context of a binary GLMM. The use of MCMC for GLMMs is particularly appealing since the conditional independencies of the model may be exploited when the required conditional distributions are calculated. Zeger and Karim (1991) described approximate Gibbs sampling for GLMMs, with nonstandar d conditional distributions being approximated by normal distributions.More general Metropolis–Hastings algorithms are straightforward to construct (see, e. g. Clayton, 1996; Gamerman, 1997). The winBUGS (Spiegelhalter, Thomas, and Best, 1998) software example manuals contain many GLMM examples. There are now a variety of additional software platforms for fitting GLMMs via MCMC including JAGS (Plummer, 2009) and BayesX (Fahrmeir and others, 2004). A large practical impediment to data analysis using MCMC is the large computational burden. For this reason, we now briefly review the INLA computational approach upon which we concentrate.The method combines Laplace approximations and numerical integration in a very efficient manner (see Rue and others, 2009, for a more extensive treatment). For the GLMM described in Section 2, the posterior is given by m Downloaded from http://biostatistics. oxfordjournals. org/ at Cornell University Library on April 20, 2013 ? y ? ? ? ?(? , ? |y ) ? ?(? |? )? (? ) i=1 y ? p(y i |? , ? ) m i=1 1 ? ? Q ? ? b ? ?(? )? (? )|Q (? 2 )|1/2 exp ? b T Q (? 2 )b + 2 y ? log p(y i |? , ? 1 ) , where y i = (yi1 , . . . , yin i ) is the vector of observations on unit/cluster i.We wish to obtain the posterior y y marginals ? (? g |y ), g = 1, . . . , G, and ? (? v |y ), v = 1, . . . , V . The number of variance components, V , should not be too large for accurate inference (since these components are integrated out via Cartesian product numerical integration, which does not scale well with dimension). We write y ? (? g |y ) = which may be evaluated via the approximation y ? (? g |y ) = K ? ? y ? ?(? g |? , y ) ? ?(? |y )d? , ? ? y ? ?(? g |? , y ) ? ? (? |y )d? ? y ? ? (? g |? k , y ) ? ? (? k |y ) ? k, ? (3. 1) k=1 here Laplace (or other related analytical approximations) are applied to carry out the integrations required ? ? for evaluation of ? (? g |? , y ). To produce the grid of points {? k , k = 1, . . . , K } over which numerical inte? y gration is performed, the mode of ? (? |y ) is located, and the Hessian is approximated, from which the grid is created and exploited in (3. 1). The output of INLA consists of posterior marginal distributions, which can be summarized via means, variances, and quantiles. Importantly for model comparison, the normaly izing constant p(y ) is calculated.The evaluation of this quantity is not straightforward using MCMC (DiCiccio and others, 1997; Meng and Wong, 1996). The deviance information criterion (Spiegelhalter, Best, and others, 1998) is popular as a model selection tool, but in random-effects models, the implicit approximation in its use is valid only when the effective number of parameters is much smaller than the number of independent observations (see Plummer, 2008). 400 Y. F ONG AND OTHERS 4. P RIOR DISTRIBUTIONS 4. 1 Fixed effects Recall that we assume ? is normally distributed. Often there will be sufficient information in the data for ? o be well estimated with a n ormal prior with a large variance (of course there will be circumstances under which we would like to specify more informative priors, e. g. when there are many correlated covariates). The use of an improper prior for ? will often lead to a proper posterior though care should be taken. For example, Wakefield (2007) shows that a Poisson likelihood with a linear link can lead to an improper posterior if an improper prior is used. Hobert and Casella (1996) discuss the use of improper priors in linear mixed effects models.If we wish to use informative priors, we may specify independent normal priors with the parameters for each component being obtained via specification of 2 quantiles with associated probabilities. For logistic and log-linear models, these quantiles may be given on the exponentiated scale since these are more interpretable (as the odds ratio and rate ratio, respectively). If ? 1 and ? 2 are the quantiles on the exponentiated scale and p1 and p2 are the associated probab ilities, then the parameters of the normal prior are given by ? = ? = z 2 log(? 1 ) ? z 1 log(? 2 ) , z2 ? 1 Downloaded from http://biostatistics. oxfordjournals. org/ at Cornell University Library on April 20, 2013 log(? 2 ) ? log(? 1 ) , z2 ? z1 where z 1 and z 2 are the p1 and p2 quantiles of a standard normal random variable. For example, in an epidemiological context, we may wish to specify a prior on a relative risk parameter, exp(? 1 ), which has a median of 1 and a 95% point of 3 (if we think it is unlikely that the relative risk associated with a unit increase in exposure exceeds 3). These specifications lead to ? 1 ? N (0, 0. 6682 ). 4. 2 Variance componentsWe begin by describing an approach for choosing a prior for a single random effect, based on Wakefield (2009). The basic idea is to specify a range for the more interpretable marginal distribution of bi and use this to drive specification of prior parameters. We state a trivial lemma upon which prior specification is ba sed, but first define some notation. We write ? ? Ga(a1 , a2 ) for the gamma distribution with un? normalized density ? a1 ? 1 exp(? a2 ? ). For q-dimensional x , we write x ? Tq (? , , d) for the Student’s x x t distribution with unnormalized density [1 + (x ? ? )T ? 1 (x ? )/d]? (d+q)/2 . This distribution has location ? , scale matrix , and degrees of freedom d. L EMMA 1 Let b|? ? N (0, ? ?1 ) and ? ? Ga(a1 , a2 ). Integration over ? gives the marginal distribution of b as T1 (0, a2 /a1 , 2a1 ). To decide upon a prior, we give a range for a generic random effect b and specify the degrees of freev d dom, d, and then solve for a1 and a2 . For the range (? R, R), we use the relationship  ±t1? (1? q)/2 a2 /a1 = d  ±R, where tq is the 100 ? qth quantile of a Student t random variable with d degrees of freedom, to give d a1 = d/2 and a2 = R 2 d/2(t1? (1? q)/2 )2 .In the linear mixed effects model, b is directly interpretable, while for binomial or Poisson models, it is more appropriate to think in terms of the marginal distribution of exp(b), the residual odds and rate ratio, respectively, and this distribution is log Student’s t. For example, if we choose d = 1 (to give a Cauchy marginal) and a 95% range of [0. 1, 10], we take R = log 10 and obtain a = 0. 5 and b = 0. 0164. Bayesian GLMMs 401 ?1 Another convenient choice is d = 2 to give the exponential distribution with mean a2 for ? ?2 . This leads to closed-form expressions for the more interpretable quantiles of ? o that, for example, if we 2 specify the median for ? as ? m , we obtain a2 = ? m log 2. Unfortunately, the use of Ga( , ) priors has become popular as a prior for ? ?2 in a GLMM context, arising from their use in the winBUGS examples manual. As has been pointed out many times (e. g. Kelsall and Wakefield, 1999; Gelman, 2006; Crainiceanu and others, 2008), this choice places the majority of the prior mass away from zero and leads to a marginal prior for the random effects which is Student’s t with 2 degrees of freedom (so that the tails are much heavier than even a Cauchy) and difficult to justify in any practical setting.We now specify another trivial lemma, but first establish notation for the Wishart distribution. For the q ? q nonsingular matrix z , we write z ? Wishartq (r, S ) for the Wishart distribution with unnormalized Downloaded from http://biostatistics. oxfordjournals. org/ at Cornell University Library on April 20, 2013 Q Lemma: Let b = (b1 , . . . , bq ), with b |Q ? iid Nq (0, Q ? 1 ), Q ? Wishartq (r, S ). Integration over Q b as Tq (0, [(r ? q + 1)S ]? 1 , r ? q + 1). S gives the marginal distribution of The margins of a multivariate Student’s t are t also, which allows r and S to be chosen as in the univariate case.Specifically, the kth element of a generic random effect, bk , follows a univariate Student t distribution with location 0, scale S kk /(r ? q + 1), and degrees of freedom d = r ? q + 1, where S kk d is element (k, k) of the inverse of S . We obtain r = d + q ? 1 and S kk = (t1? (1? q)/2 )2 /(d R 2 ). If a priori b are correlated we may specify S jk = 0 for j = k and we have no reason to believe that elements of S kk = 1/Skk , to recover the univariate specification, recognizing that with q = 1, the univariate Wishart has parameters a1 = r/2 and a2 = 1/(2S).If we believe that elements of b are dependent then we may specify the correlations and solve for the off-diagonal elements of S . To ensure propriety of the posterior, proper priors are required for ; Zeger and Karim (1991) use an improper prior for , so that the posterior is improper also. 4. 3 Effective degrees of freedom variance components prior z z z z density |z |(r ? q? 1)/2 exp ? 1 tr(z S ? 1 ) . This distribution has E[z ] = r S and E[z ? 1 ] = S ? 1 /(r ? q ? 1), 2 and we require r > q ? 1 for a proper distribution.In Section 5. 3, we describe the GLMM representation of a spline model. A generic linear spline model is given by K yi = x i ? + k=1 z ik bk + i , where x i is a p ? 1 vector of covariates with p ? 1 associated fixed effects ? , z ik denote the spline 2 basis, bk ? iid N (0, ? b ), and i ? iid N (0, ? 2 ), with bk and i independent. Specification of a prior for 2 is not straightforward, but may be of great importance since it contributes to determining the amount ? b of smoothing that is applied. Ruppert and others (2003, p. 77) raise concerns, â€Å"about the instability of automatic smoothing parameter selection even for single predictor models†, and continue, â€Å"Although we are attracted by the automatic nature of the mixed model-REML approach to fitting additive models, we discourage blind acceptance of whatever answer it provides and recommend looking at other amounts of smoothing†. While we would echo this general advice, we believe that a Bayesian mixed model approach, with carefully chosen priors, can increase the stability of the mixed model representation. There has be en 2 some discussion of choice of prior for ? in a spline context (Crainiceanu and others, 2005, 2008). More general discussion can be found in Natarajan and Kass (2000) and Gelman (2006). In practice (e. g. Hastie and Tibshirani, 1990), smoothers are often applied with a fixed degrees of freedom. We extend this rationale by examining the prior degrees of freedom that is implied by the choice 402 Y. F ONG AND OTHERS ?2 ? b ? Ga(a1 , a2 ). For the general linear mixed model y = x ? + zb + , we have x z where C = [x |z ] is n ? ( p + K ) and C y = x ? + z b = C (C T C + 0 p? p 0K ? p )? 1 C T y , = 0 p? K 2 cov(b )? 1 b ? )? 1 C T C }, Downloaded from http://biostatistics. xfordjournals. org/ at Cornell University Library on April 20, 2013 (see, e. g. Ruppert and others, 2003, Section 8. 3). The total degrees of freedom associated with the model is C df = tr{(C T C + which may be decomposed into the degrees of freedom associated with ? and b , and extends easily to situations in which we have additional random effects, beyond those associated with the spline basis (such an example is considered in Section 5. 3). In each of these situations, the degrees of freedom associated C with the respective parameter is obtained by summing the appropriate diagonal elements of (C T C + )? C T C . Specifically, if we have j = 1, . . . , d sets of random-effect parameters (there are d = 2 in the model considered in Section 5. 3) then let E j be the ( p + K ) ? ( p + K ) diagonal matrix with ones in the diagonal positions corresponding to set j. Then the degrees of freedom associated with this set is E C df j = tr{E j (C T C + )? 1 C T C . Note that the effective degrees of freedom changes as a function of K , as expected. To evaluate , ? 2 is required. If we specify a proper prior for ? 2 , then we may specify the 2 2 joint prior as ? (? b , ? 2 ) = ? (? 2 )? (? b |? 2 ).Often, however, we assume the improper prior ? (? 2 ) ? 1/? 2 since the data provide sufficient information with respect to ? 2 . Hence, we have found the substitution of an estimate for ? 2 (for example, from the fitting of a spline model in a likelihood implementation) to be a practically reasonable strategy. As a simple nonspline demonstration of the derived effective degrees of freedom, consider a 1-way analysis of variance model Yi j = ? 0 + bi + i j 2 with bi ? iid N (0, ? b ), i j ? iid N (0, ? 2 ) for i = 1, . . . , m = 10 groups and j = 1, . . . , n = 5 observa? 2 tions per group. For illustration, we assume ? ? Ga(0. 5, 0. 005). Figure 1 displays the prior distribution for ? , the implied prior distribution on the effective degrees of freedom, and the bivariate plot of these quantities. For clarity of plotting, we exclude a small number of points beyond ? > 2. 5 (4% of points). In panel (c), we have placed dashed horizontal lines at effective degrees of freedom equal to 1 (complete smoothing) and 10 (no smoothing). From panel (b), we conclude that here the prior choice favors q uite strong smoothing. This may be contrasted with the gamma prior with parameters (0. 001, 0. 001), which, in this example, gives reater than 99% of the prior mass on an effective degrees of freedom greater than 9. 9, again showing the inappropriateness of this prior. It is appealing to extend the above argument to nonlinear models but unfortunately this is not straightforward. For a nonlinear model, the degrees of freedom may be approximated by C df = tr{(C T W C + where W = diag Vi? 1 d? i dh 2 )? 1 C T W C }, and h = g ? 1 denotes the inverse link function. Unfortunately, this quantity depends on ? and b , which means that in practice, we would have to use prior estimates for all of the parameters, which may not be practically possible.Fitting the model using likelihood and then substituting in estimates for ? and b seems philosophically dubious. Bayesian GLMMs 403 Downloaded from http://biostatistics. oxfordjournals. org/ at Cornell University Library on April 20, 2013 Fig. 1. Gamma prior for ? ?2 with parameters 0. 5 and 0. 005, (a) implied prior for ? , (b) implied prior for the effective degrees of freedom, and (c) effective degrees of freedom versus ? . 4. 4 Random walk models Conditionally represented smoothing models are popular for random effects in both temporal and spatial applications (see, e. g. Besag and others, 1995; Rue and Held, 2005).For illustration, consider models of the form ? (m? r ) Q u 2 exp ? p(u |? u ) = (2? )? (m? r )/2 |Q |1/2 ? u 1 T u Qu , 2 2? u (4. 1) 404 Y. F ONG AND OTHERS where u = (u 1 , . . . , u m ) is the collection of random effects, Q is a (scaled) â€Å"precision† matrix of rank Q m ? r , whose form is determined by the application at hand, and |Q | is a generalized determinant which is the product over the m ? r nonzero eigenvalues of Q . Picking a prior for ? u is not straightforward because ? u has an interpretation as the conditional standard deviation, where the elements that are conditioned upon depend s on the application.We may simulate realizations from (4. 1) to examine candidate prior distributions. Due to the rank deficiency, (4. 1) does not define a probability density, and so we cannot directly simulate from this prior. However, Rue and Held (2005) give an algorithm for generating samples from (4. 1): 1. Simulate z j ? N (0, 1 ), for j = m ? r + 1, . . . , m, where ? j are the eigenvalues of Q (there are j m ? r nonzero eigenvalues as Q has rank m ? r ). 2. Return u = z m? r +1 e n? r +1 + z 3 e 3 + †¢ †¢ †¢ + z n e m = E z , where e j are the corresponding eigenvectors of Q , E is the m ? (m ? ) matrix with these eigenvectors as columns, and z is the (m ? r ) ? 1 vector containing z j , j = m ? r + 1, . . . , m. The simulation algorithm is conditioned so that samples are zero in the null-space of Q ; if u is a sample and the null-space is spanned by v 1 and v 2 , then u T v 1 = u T v 2 = 0. For example, suppose Q 1 = 0 so that the null-space is spanned by 1, and the rank deficiency is 1. Then Q is improper since the eigenvalue corresponding to 1 is zero, and samples u produced by the algorithm are such that u T 1 = 0. In Section 5. 2, we use this algorithm to evaluate different priors via simulation.It is also useful to note that if we wish to compute the marginal variances only, simulation is not required, as they are available as the diagonal elements of the matrix j 1 e j e T . j j 5. E XAMPLES Here, we report 3 examples, with 4 others described in the supplementary material available at Biostatistics online. Together these cover all the examples in Breslow and Clayton (1993), along with an additional spline example. In the first example, results using the INLA numerical/analytical approximation described in Section 3 were compared with MCMC as implemented in the JAGS software (Plummer, 2009) and found to be accurate.For the models considered in the second and third examples, the approximation was compared with the MCMC implement ation contained in the INLA software. 5. 1 Longitudinal data We consider the much analyzed epilepsy data set of Thall and Vail (1990). These data concern the number ? of seizures, Yi j for patient i on visit j, with Yi j |? , b i ? ind Poisson(? i j ), i = 1, . . . , 59, j = 1, . . . , 4. We concentrate on the 3 random-effects models fitted by Breslow and Clayton (1993): log ? i j = x i j ? + b1i , (5. 1) (5. 2) (5. 3) Downloaded from http://biostatistics. oxfordjournals. rg/ at Cornell University Library on April 20, 2013 log ? i j = x i j ? + b1i + b2i V j /10, log ? i j = x i j ? + b1i + b0i j , where x i j is a 1 ? 6 vector containing a 1 (representing the intercept), an indicator for baseline measurement, a treatment indicator, the baseline by treatment interaction, which is the parameter of interest, age, and either an indicator of the fourth visit (models (5. 1) and (5. 2) and denoted V4 ) or visit number coded ? 3, ? 1, +1, +3 (model (5. 3) and denoted V j /10) and ? is the associated fixed effect. All 3 models 2 include patient-specific random effects b1i ? N 0, ? , while in model (5. 2), we introduce independent 2 ). Model (5. 3) includes random effects on the slope associated with â€Å"measurement errors,† b0i j ? N (0, ? 0 Bayesian GLMMs 405 Table 1. PQL and INLA summaries for the epilepsy data Variable Base Trt Base ? Trt Age V4 or V/10 ? 0 ? 1 ? 2 Model (5. 1) PQL 0. 87  ± 0. 14 ? 0. 91  ± 0. 41 0. 33  ± 0. 21 0. 47  ± 0. 36 ? 0. 16  ± 0. 05 — 0. 53  ± 0. 06 — INLA 0. 88  ± 0. 15 ? 0. 94  ± 0. 44 0. 34  ± 0. 22 0. 47  ± 0. 38 ? 0. 16  ± 0. 05 — 0. 56  ± 0. 08 — Model (5. 2) PQL 0. 86  ± 0. 13 ? 0. 93  ± 0. 40 0. 34  ± 0. 21 0. 47  ± 0. 35 ? 0. 10  ± 0. 09 0. 36  ± 0. 04 0. 48  ± 0. 06 — INLA 0. 8  ± 0. 15 ? 0. 96  ± 0. 44 0. 35  ± 0. 23 0. 48  ± 0. 39 ? 0. 10  ± 0. 09 0. 41  ± 0. 04 0. 53  ± 0. 07 — Model (5. 3) PQL 0. 87  ± 0. 14 ? 0. 91  ± 0. 41 0. 33  ± 0. 21 0. 46  ± 0. 36 ? 0. 26  ± 0. 16 — 0. 52  ± 0. 06 0. 74  ± 0. 16 INLA 0. 88  ± 0. 14 ? 0. 94  ± 0. 44 0. 34  ± 0. 22 0. 47  ± 0. 38 ? 0. 27  ± 0. 16 — 0. 56  ± 0. 06 0. 70  ± 0. 14 Downloaded from http://biostatistics. oxfordjournals. org/ at Cornell University Library on April 20, 2013 visit, b2i with b1i b2i ? N (0, Q ? 1 ). (5. 4) We assume Q ? Wishart(r, S ) with S = S11 S12 . For prior specification, we begin with the bivariate S21 S22 model and assume that S is diagonal.We assume the upper 95% point of the priors for exp(b1i ) and exp(b2i ) are 5 and 4, respectively, and that the marginal distributions are t with 4 degrees of freedom. Following the procedure outlined in Section 4. 2, we obtain r = 5 and S = diag(0. 439, 0. 591). We take ? 2 the prior for ? 1 in model (5. 1) to be Ga(a1 , a2 ) with a1 = (r ? 1)/2 = 2 and a2 = 1/2S11 = 1. 140 (so that this prior coincides with the marginal prior obtained from the bivariat e specification). In model (5. 2), ? 2 ? 2 we assume b1i and b0i j are independent, and that ? 0 follows the same prior as ? , that is, Ga(2, 1. 140). We assume a flat prior on the intercept, and assume that the rate ratios, exp(? j ), j = 1, . . . , 5, lie between 0. 1 and 10 with probability 0. 95 which gives, using the approach described in Section 4. 1, a normal prior with mean 0 and variance 1. 172 . Table 1 gives PQL and INLA summaries for models (5. 1–5. 3). There are some differences between the PQL and Bayesian analyses, with slightly larger standard deviations under the latter, which probably reflects that with m = 59 clusters, a little accuracy is lost when using asymptotic inference.There are some differences in the point estimates which is at least partly due to the nonflat priors used—the priors have relatively large variances, but here the data are not so abundant so there is sensitivity to the prior. Reassuringly under all 3 models inference for the bas eline-treatment interaction of interest is virtually y identical and suggests no significant treatment effect. We may compare models using log p(y ): for 3 models, we obtain values of ? 674. 8, ? 638. 9, and ? 665. 5, so that the second model is strongly preferred. 5. Smoothing of birth cohort effects in an age-cohort model We analyze data from Breslow and Day (1975) on breast cancer rates in Iceland. Let Y jk be the number of breast cancer of cases in age group j (20–24,. . . , 80–84) and birth cohort k (1840–1849,. . . ,1940–1949) with j = 1, . . . , J = 13 and k = 1, . . . , K = 11. Following Breslow and Clayton (1993), we assume Y jk |? jk ? ind Poisson(? jk ) with log ? jk = log n jk + ? j + ? k + vk + u k (5. 5) and where n jk is the person-years denominator, exp(? j ), j = 1, . . . , J , represent fixed effects for age relative risks, exp(? is the relative risk associated with a one group increase in cohort group, vk ? iid 406 Y. F ONG AND OTHERS 2 N (0, ? v ) represent unstructured random effects associated with cohort k, with smooth cohort terms u k following a second-order random-effects model with E[u k |{u i : i < k}] = 2u k? 1 ? u k? 2 and Var(u k |{u i : 2 i < k}) = ? u . This latter model is to allow the rates to vary smoothly with cohort. An equivalent representation of this model is, for 2 < k < K ? 1, 1 E[u k |{u l : l = k}] = (4u k? 1 + 4u k+1 ? u k? 2 ? u k+2 ), 6 Var(u k |{u l : l = k}) = 2 ? . 6 Downloaded from http://biostatistics. oxfordjournals. org/ at Cornell University Library on April 20, 2013 The rank of Q in the (4. 1) representation of this model is K ? 2 reflecting that both the overall level and the overall trend are aliased (hence the appearance of ? in (5. 5)). The term exp(vk ) reflects the unstructured residual relative risk and, following the argument in Section 4. 2, we specify that this quantity should lie in [0. 5, 2. 0] with probability 0. 95, with a marginal log Cauchy ? 2 distribution, to obtain the gamma prior ? v ? Ga(0. 5, 0. 00149).The term exp(u k ) reflects the smooth component of the residual relative risk, and the specification of a 2 prior for the associated variance component ? u is more difficult, given its conditional interpretation. Using the algorithm described in Section 4. 2, we examined simulations of u for different choices of gamma ? 2 hyperparameters and decided on the choice ? u ? Ga(0. 5, 0. 001); Figure 2 shows 10 realizations from the prior. The rationale here is to examine realizations to see if they conform to our prior expectations and in particular exhibit the required amount of smoothing.All but one of the realizations vary smoothly across the 11 cohorts, as is desirable. Due to the tail of the gamma distribution, we will always have some extreme realizations. The INLA results, summarized in graphical form, are presented in Figure 2(b), alongside likelihood fits in which the birth cohort effect is incorporated as a linear term and as a f actor. We see that the smoothing model provides a smooth fit in birth cohort, as we would hope. 5. 3 B-Spline nonparametric regression We demonstrate the use of INLA for nonparametric smoothing using O’Sullivan splines, which are based on a B-spline basis.We illustrate using data from Bachrach and others (1999) that concerns longitudinal measurements of spinal bone mineral density (SBMD) on 230 female subjects aged between 8 and 27, and of 1 of 4 ethnic groups: Asian, Black, Hispanic, and White. Let yi j denote the SBMD measure for subject i at occasion j, for i = 1, . . . , 230 and j = 1, . . . , n i with n i being between 1 and 4. Figure 3 shows these data, with the gray lines indicating measurements on the same woman. We assume the model K Yi j = x i ? 1 + agei j ? 2 + k=1 z i jk b1k + b2i + ij, where x i is a 1 ? vector containing an indicator for the ethnicity of individual i, with ? 1 the associated 4 ? 1 vector of fixed effects, z i jk is the kth basis associated with age, with associated parameter b1k ? 2 2 N (0, ? 1 ), and b2i ? N (0, ? 2 ) are woman-specific random effects, finally, i j ? iid N (0, ? 2 ). All random terms are assumed independent. Note that the spline model is assumed common to all ethnic groups and all women, though it would be straightforward to allow a different spline for each ethnicity. Writing this model in the form y = x ? + z 1b1 + z 2b 2 + = C ? + . Bayesian GLMMs 407Downloaded from http://biostatistics. oxfordjournals. org/ at Cornell University Library on April 20, 2013 Fig. 2. (a) Ten realizations (on the relative risk scale) from the random effects second-order random walk model in which the prior on the random-effects precision is Ga(0. 5,0. 001), (b) summaries of fitted models: the solid line corresponds to a log-linear model in birth cohort, the circles to birth cohort as a factor, and â€Å"+† to the Bayesian smoothing model. we use the method described in Section 4. 3 to examine the effective number of parameters implied by the ? 2 ? 2 priors ? 1 ? Ga(a1 , a2 ) and ? 2 ? Ga(a3 , a4 ).To fit the model, we first use the R code provided in Wand and Ormerod (2008) to construct the basis functions, which are then input to the INLA program. Running the REML version of the model, we obtain 2 ? = 0. 033 which we use to evaluate the effective degrees of freedoms associated with priors for ? 1 and 2 . We assume the usual improper prior, ? (? 2 ) ? 1/? 2 for ? 2 . After some experimentation, we settled ? 2 408 Y. F ONG AND OTHERS Downloaded from http://biostatistics. oxfordjournals. org/ at Cornell University Library on April 20, 2013 Fig. 3. SBMD versus age by ethnicity. Measurements on the same woman are joined with gray lines.The solid curve corresponds to the fitted spline and the dashed lines to the individual fits. ?2 2 on the prior ? 1 ? Ga(0. 5, 5 ? 10? 6 ). For ? 2 , we wished to have a 90% interval for b2i of  ±0. 3 which, ? 2 with 1 degree of freedom for the marginal distributio n, leads to ? 2 ? Ga(0. 5, 0. 00113). Figure 4 shows the priors for ? 1 and ? 2 , along with the implied effective degrees of freedom under the assumed priors. For the spline component, the 90% prior interval for the effective degrees of freedom is [2. 4,10]. Table 2 compares estimates from REML and INLA implementations of the model, and we see close correspondence between the 2.Figure 4 also shows the posterior medians for ? 1 and ? 2 and for the 2 effective degrees of freedom. For the spline and random effects these correspond to 8 and 214, respectively. The latter figure shows that there is considerable variability between the 230 women here. This is confirmed in Figure 3 where we observe large vertical differences between the profiles. This figure also shows the fitted spline, which appears to mimic the trend in the data well. 5. 4 Timings For the 3 models in the longitudinal data example, INLA takes 1 to 2 s to run, using a single CPU.To get estimates with similar precision wit h MCMC, we ran JAGS for 100 000 iterations, which took 4 to 6 min. For the model in the temporal smoothing example, INLA takes 45 s to run, using 1 CPU. Part of the INLA procedure can be executed in a parallel manner. If there are 2 CPUs available, as is the case with today’s prevalent INTEL Core 2 Duo processors, INLA only takes 27 s to run. It is not currently possible to implement this model in JAGS. We ran the MCMC utility built into the INLA software for 3. 6 million iterations, to obtain estimates of comparable accuracy, which took 15 h.For the model in the B-spline nonparametric regression example, INLA took 5 s to run, using a single CPU. We ran the MCMC utility built into the INLA software for 2. 5 million iterations to obtain estimates of comparable accuracy, the analysis taking 40 h. Bayesian GLMMs 409 Downloaded from http://biostatistics. oxfordjournals. org/ at Cornell University Library on April 20, 2013 Fig. 4. Prior summaries: (a) ? 1 , the standard deviation of the spline coefficients, (b) effective degrees of freedom associated with the prior for the spline coefficients, (c) effective degrees of freedom versus ? , (d) ? 2 , the standard deviation of the between-individual random effects, (e) effective degrees of freedom associated with the individual random effects, and (f) effective degrees of freedom versus ? 2 . The vertical dashed lines on panels (a), (b), (d), and (e) correspond to the posterior medians. Table 2. REML and INLA summaries for spinal bone data. Intercept corresponds to Asian group Variable Intercept Black Hispanic White Age ? 1 ? 2 ? REML 0. 560  ± 0. 029 0. 106  ± 0. 021 0. 013  ± 0. 022 0. 026  ± 0. 022 0. 021  ± 0. 002 0. 018 0. 109 0. 033 INLA 0. 563  ± 0. 031 0. 106  ± 0. 021 0. 13  ± 0. 022 0. 026  ± 0. 022 0. 021  ± 0. 002 0. 024  ± 0. 006 0. 109  ± 0. 006 0. 033  ± 0. 002 Note: For the entries marked with a standard errors were unavailable. 410 Y. F ONG AND OTHERS 6. D ISCUSSION In t his paper, we have demonstrated the use of the INLA computational method for GLMMs. We have found that the approximation strategy employed by INLA is accurate in general, but less accurate for binomial data with small denominators. The supplementary material available at Biostatistics online contains an extensive simulation study, replicating that presented in Breslow and Clayton (1993).There are some suggestions in the discussion of Rue and others (2009) on how to construct an improved Gaussian approximation that does not use the mode and the curvature at the mode. It is likely that these suggestions will improve the results for binomial data with small denominators. There is an urgent need for diagnosis tools to flag when INLA is inaccurate. Conceptually, computation for nonlinear mixed effects models (Davidian and Giltinan, 1995; Pinheiro and Bates, 2000) can also be handled by INLA but this capability is not currently available. The website www. r-inla. rg contains all the data and R scripts to perform the analyses and simulations reported in the paper. The latest release of software to implement INLA can also be found at this site. Recently, Breslow (2005) revisited PQL and concluded that, â€Å"PQL still performs remarkably well in comparison with more elaborate procedures in many practical situations. † We believe that INLA provides an attractive alternative to PQL for GLMMs, and we hope that this paper stimulates the greater use of Bayesian methods for this class. Downloaded from http://biostatistics. oxfordjournals. org/ at Cornell University Library on April 20, 2013S UPPLEMENTARY MATERIAL Supplementary material is available at http://biostatistics. oxfordjournals. org. ACKNOWLEDGMENT Conflict of Interest: None declared. F UNDING National Institutes of Health (R01 CA095994) to J. W. Statistics for Innovation (sfi. nr. no) to H. R. R EFERENCES BACHRACH , L. K. , H ASTIE , T. , WANG , M. C. , NARASIMHAN , B. AND M ARCUS , R. (1999). Bone mineral acquisition in healthy Asian, Hispanic, Black and Caucasian youth. A longitudinal study. The Journal of Clinical Endocrinology and Metabolism 84, 4702–4712. B ESAG , J. , G REEN , P. J. , H IGDON , D. AND M ENGERSEN , K. 1995). Bayesian computation and stochastic systems (with discussion). Statistical Science 10, 3–66. B RESLOW, N. E. (2005). Whither PQL? In: Lin, D. and Heagerty, P. J. (editors), Proceedings of the Second Seattle Symposium. New York: Springer, pp. 1–22. B RESLOW, N. E. AND C LAYTON , D. G. (1993). Approximate inference in generalized linear mixed models. Journal of the American Statistical Association 88, 9–25. B RESLOW, N. E. AND DAY, N. E. (1975). Indirect standardization and multiplicative models for rates, with reference to the age adjustment of cancer incidence and relative frequency data.Journal of Chronic Diseases 28, 289–301. C LAYTON , D. G. (1996). Generalized linear mixed models. In: Gilks, W. R. , Richardson, S. and S piegelhalter, D. J. (editors), Markov Chain Monte Carlo in Practice. London: Chapman and Hall, pp. 275–301. Bayesian GLMMs 411 C RAINICEANU , C. M. , D IGGLE , P. J. AND ROWLINGSON , B. (2008). Bayesian analysis for penalized spline regression using winBUGS. Journal of the American Statistical Association 102, 21–37. C RAINICEANU , C. M. , RUPPERT, D. AND WAND , M. P. (2005). Bayesian analysis for penalized spline regression using winBUGS. Journal of Statistical Software 14.DAVIDIAN , M. AND G ILTINAN , D. M. (1995). Nonlinear Models for Repeated Measurement Data. London: Chapman and Hall. D I C ICCIO , T. J. , K ASS , R. E. , R AFTERY, A. AND WASSERMAN , L. (1997). Computing Bayes factors by combining simulation and asymptotic approximations. Journal of the American Statistical Association 92, 903–915. Downloaded from http://biostatistics. oxfordjournals. org/ at Cornell University Library on April 20, 2013 D IGGLE , P. , H EAGERTY, P. , L IANG , K. -Y. Oxford: Oxford University Press. AND Z EGER , S. (2002). Analysis of Longitudinal Data, 2nd edition. FAHRMEIR , L. , K NEIB , T.AND L ANG , S. (2004). Penalized structured additive regression for space-time data: a Bayesian perspective. Statistica Sinica 14, 715–745. G AMERMAN , D. (1997). Sampling from the posterior distribution in generalized linear mixed models. Statistics and Computing 7, 57–68. G ELMAN , A. (2006). Prior distributions for variance parameters in hierarchical models. Bayesian Analysis 1, 515–534. H ASTIE , T. J. AND T IBSHIRANI , R. J. (1990). Generalized Additive Models. London: Chapman and Hall. H OBERT, J. P. AND C ASELLA , G. (1996). The effect of improper priors on Gibbs sampling in hierarchical linear mixed models.Journal of the American Statistical Association 91, 1461–1473. K ASS , R. E. AND S TEFFEY, D. (1989). Approximate Bayesian inference in conditionally independent hierarchical models (parametric empirical Bayes models). Journal of the American Statistical Association 84, 717–726. K ELSALL , J. E. AND WAKEFIELD , J. C. (1999). Discussion of â€Å"Bayesian models for spatially correlated disease and exposure data† by N. Best, I. Waller, A. Thomas, E. Conlon and R. Arnold. In: Bernardo, J. M. , Berger, J. O. , Dawid, A. P. and Smith, A. F. M. (editors), Sixth Valencia International Meeting on Bayesian Statistics. London: Oxford University Press.M C C ULLAGH , P. AND N ELDER , J. A. (1989). Generalized Linear Models, 2nd edition. London: Chapman and Hall. M C C ULLOCH , C. E. , S EARLE , S. R. AND N EUHAUS , J. M. (2008). Generalized, Linear, and Mixed Models, 2nd edition. New York: John Wiley and Sons. M ENG , X. AND W ONG , W. (1996). Simulating ratios of normalizing constants via a simple identity. Statistical Sinica 6, 831–860. NATARAJAN , R. AND K ASS , R. E. (2000). Reference Bayesian methods for generalized linear mixed models. Journal of the American Statistical Association 95, 22 7–237. N ELDER , J. AND W EDDERBURN , R. (1972). Generalized linear models.Journal of the Royal Statistical Society, Series A 135, 370–384. P INHEIRO , J. C. AND BATES , D. M. (2000). Mixed-Effects Models in S and S-plus. New York: Springer. P LUMMER , M. (2008). Penalized loss functions for Bayesian model comparison. Biostatistics 9, 523–539. P LUMMER , M. (2009). Jags version 1. 0. 3 manual. Technical Report. RUE , H. AND H ELD , L. (2005). Gaussian Markov Random Fields: Thoery and Application. Boca Raton: Chapman and Hall/CRC. RUE , H. , M ARTINO , S. AND C HOPIN , N. (2009). Approximate Bayesian inference for latent Gaussian models using integrated nested laplace approximations (with discussion).Journal of the Royal Statistical Society, Series B 71, 319–392. 412 RUPPERT, D. R. , WAND , M. P. University Press. AND Y. F ONG AND OTHERS C ARROLL , R. J. (2003). Semiparametric Regression. New York: Cambridge S KENE , A. M. AND WAKEFIELD , J. C. (1990). Hie rarchical models for multi-centre binary response studies. Statistics in Medicine 9, 919–929. S PIEGELHALTER , D. , B EST, N. , C ARLIN , B. AND VAN DER L INDE , A. (1998). Bayesian measures of model complexity and fit (with discussion). Journal of the Royal Statistical Society, Series B 64, 583–639. S PIEGELHALTER , D. J. , T HOMAS , A.AND B EST, N. G. (1998). WinBUGS User Manual. Version 1. 1. 1. Cambridge. T HALL , P. F. AND VAIL , S. C. (1990). Some covariance models for longitudinal count data with overdispersion. Biometrics 46, 657–671. V ERBEKE , G. V ERBEKE , G. AND AND Downloaded from http://biostatistics. oxfordjournals. org/ at Cornell University Library on April 20, 2013 M OLENBERGHS , G. (2000). Linear Mixed Models for Longitudinal Data. New York: Springer. M OLENBERGHS , G. (2005). Models for Discrete Longitudinal Data. New York: Springer. WAKEFIELD , J. C. (2007). Disease mapping and spatial regression with count data.Biostatistics 8, 158–1 83. WAKEFIELD , J. C. (2009). Multi-level modelling, the ecologic fallacy, and hybrid study designs. International Journal of Epidemiology 38, 330–336. WAND , M. P. AND O RMEROD , J. T. (2008). On semiparametric regression with O’Sullivan penalised splines. Australian and New Zealand Journal of Statistics 50, 179–198. Z EGER , S. L. AND K ARIM , M. R. (1991). Generalized linear models with random effects: a Gibbs sampling approach. Journal of the American Statistical Association 86, 79–86. [Received September 4, 2009; revised November 4, 2009; accepted for publication November 6, 2009]

Cost benefit analysis report Coursework Example | Topics and Well Written Essays - 2500 words

Cost benefit analysis report - Coursework Example By free wifi services, it means that the public should be in a position to access internet service while travelling. Since a great percent of the nation is learned, the government should come up with measures to help them access internet services. This will act in a way to improve knowledge in various market centers for those in business, schools or those planning to do business. The main situation on ground, is that majority of people in the country begin their days early to work and use public service vehicles. Many are usually bored because they all meet as strangers and find it difficult to start a conversation so as to break monotony and end up silent till they alight from the vehicles. The introduction of wifi will make passengers enjoy their commutation to work. Most people who use internet services are in business or in school. They need to discover new ways of increasing productivity and try beating the upcoming competition especially for those in business. Free wifi will allow most people after closing their businesses at the end of the day, be able to get new tips of improving their productivity on their way home. Cost benefit refers to, a systematic analysis that shows an approach in estimating the strengths and weaknesses of alternatives that make transactions, activities or functional requirements for business be considered to be of quality. Thus, David, Ngulube and Dude (2013), go ahead to explain that the technique is used to determine options that give the best approach for a business to be adopted and practiced. The benefits are looked in terms of lab our, time and saving costs. It will be practical to the free wifi that we are advocating for. In terms of time saving, the business persons will be able to get new tips on ways to improve attraction from customers without having to visit cyber cafes or areas where internet is available during their off days. It will be

Make a Difference Paper Essay Example | Topics and Well Written Essays - 500 words

Make a Difference Paper - Essay Example The easiest way is tapping the existing parents and teacher's organizations in the schools and other important personalities are to be added as long as necessary. Proper alignment of responsibilities will be identified and focus shall be on major tasks identified. These tasked includes: research and baseline data gathering, program designing and testing, implementation, and monitoring and evaluation. This step can be carried out through survey and direct evaluation. Questionnaires can be devised through assistance of experts to collect data such as: current participatory practices, what practices are effective or not, what is the extent of involvement of the parents, what are the problems encountered with current participatory programs. In other words, a thorough analysis of the weaknesses and strengths (the inner resources of those directly involved), and the opportunities and threats (the uncontrollable external factors) within the society in general must be undertaken. There are systematic means in doing this and a lot of experts can be tapped to do effective participative SWOT analysis. After examining everything under the SWOT analysis an effective program can be designed.

Working with and Leading people Essay Example | Topics and Well Written Essays - 1500 words

Working with and Leading people - Essay Example Testing and assessment should also be done in this stage. The next step will be interviewing the candidates. This allows the organization to further understand a candidate. An interview should go beyond the skills in a resume or the score of a test - it should focus on how the candidate can apply the skills and scores in the work environment. For a position of call centre are much needed, a person possessing high level of communication skills and social interaction. These agents should have the capacity to handle customer service in an international set-up. Applicants should also possess technical knowledge in the fields where the agent is assigned. For call centre manager handling corporate clients, a perfect candidate would be a person that has superb leadership and management skills. He should also display mastery in receiving/conducting calls so that he can be a perfect example to the staff he is leading. Therefore he should have effective communication and interaction skills that his subordinates can emulate. For a human resources department administrator, an applicant should have excellent presentation skills and can motivate others by his inspirational and motivating speeches. Call centre business needs someone who can uplift the spirits of agents who are prone to be upset by distracting and irritating calls. This is exactly their job. A manager A manager that has a just and conscientious vision for his fellow employees, that is what a human resources manager should possess. He should have the zeal to improve his other employees in the company. To put this vision into action, the candidate for this post should have superb managerial and planning skills and capable of spearheading the organization towards staff development. He should also have sufficient knowledge in pertinent legal knowledge needed to ensure that employees will be given just compensation and benefits. Just like the human resources manager, the senior sales manager should be competent in meticulously planning the sales campaign of the company. He should be able to produce sales targets and quotas for his sales representatives. Training his sales force is his other significant responsibility. Because of this, he should have excellent presentation and training skills (Assessment and Selection, 2007). Security staff should be vigilant and physically fit for the position. Since security personnel are needed 24 hours a day, 7 days a week, applicants should also be willing to work in different 8-hour shifts. In recruiting, human resources personnel should be wary about certain legal, regulatory and ethical requirements. The first one is the character and productivity background of the applicant. HR personnel have the responsibility to look over the attitude and behaviour of its applicants. Yet in doing so, the HR should be cautious about the legal issues in recruiting. One such concern is the right of the applicants against discrimination from disability. Disability Discrimination Act 1995 in the UK protects the rights of the disabled (Employer's Forum for Disability, 2004). Aside from this, UK has various laws that will protect employee's rights, such as Race Relations Act 1976, Sex Discrimination Act 1975, and Employment Equality Regulations 2003 (Independent Police Complaints Commission, n.d., p.2). Full-time long term employees have the experience unparalleled by the new-comers. They also have the zealous concern in

International business law Assignment Example | Topics and Well Written Essays - 3000 words - 1

International business law - Assignment Example Every citizen of a state has certain duties towards the state and in the interest of the society as a whole, the state prescribes certain norms of conduct that bind all members of the state upon which it has a jurisdiction to punish a person who transgresses against these (Kleyn & Viljoen 2002). A good example of a criminal offence is theft which is unlawful taking of something that does not belong to you. The sole purpose of theft law in this case is to prevent interference with property rights. As far as criminal law is concerned, the state is part of the proceedings against the accused person. Criminal law is seen as an effective way of dealing with certain conducts which are deemed wrongful and violate the prescribed norms in society. In the case of the United States Court of Appeals Tenth Circuit (2007), the defendant Juan Carlos Elizade has been convicted as an â€Å"aggravated felony† for joyriding and was subsequently sentenced to a one year suspended sentence. Thus, cases with the intent of temporarily depriving the owner of his property rights constitute a criminal offense. On the other hand, civil law is different from criminal law in that it constitutes private law which specifically deals with legal relationships between subjects (Kleyn & Viljoen 2002). The subjects in civil proceedings are relatively on the same footing with each other and the state only acts as an arbiter. Civil cases often involve family law, tort or contract laws. In a civil case, it is the plaintiff versus the defendant while in a criminal case it is the state versus the accused. The parties to a civil case decide whether they want to initiate proceedings while in a criminal case, the state initiates prosecution. The definite difference between a criminal and civil case is that the aim of criminal law is to punish the subjects threatening order and harmony in society while in contrast, the aim

Unit 3DB The Green Revolution Research Paper Example | Topics and Well Written Essays - 250 words

Unit 3DB The Green Revolution - Research Paper Example In this regard, energy conservation should be highly rated and considered if indeed the long term effects of environmental conservation are to be realised. In order to reduce personal energy consumption, the following changes can be made that would in the long run create a significant difference. To save car fuel energy consumed on daily drives to work, commuter trains and or bicycles can be used to achieve the same. This will not only reduce gasoline fuel consumption, but also help in environmental pollution. Concerning the other forms of energy use, such as the ones on lighting heating and cooking, renewable energy, specifically, solar and wind energy can be utilised. These energy forms are sustainable and also clean. On the other hand, to increase energy efficiency requires discipline in the use of already in place energy sources. This means small changes such as switching off unnecessary lights around the house, and using energy efficient light saver bulbs. In addition to this, electric appliances should be used as per the manufactures specifications. Household electric appliances account for nearly thirty per cent of home electric consumption. Therefore, efficient use will directly translate to efficient energy consumption. An electric vehicle consists of electric motors that are used to drive the car. These can be powered externally, through stored electric energy such as batteries, and through an internal energy source such as generators. Electric cars are considered to be efficient, clean, and most of all environment friendly. In this regard, yes I would purchase an electric vehicle because of its long term benefits. Fuel costs are bound to rise, that is guaranteed, In addition to that fossil fuel is not renewable and as such contributes to environmental degradation. Based on this alone, using an electric vehicle contributes to energy conservation, has long term cost benefits, and is also energy

Macroeconomics Bachelor Essay Example | Topics and Well Written Essays - 750 words

Macroeconomics Bachelor - Essay Example It was noted that Central Banks have been successful in their policies which controlled inflation particularly in controlling insulating countries from shocks such as high oil prices. This mechanism will be explained using various macroeconomic principles. The money supply is directly linked with inflation as shown in the famous Quantity Theory of Money (QTM). This model links the level of money supply to the level of prices of goods and services sold, thus inflation. The famous equation of the TQM is MV = PT, where P is the average price level, T is the volume of transactions of goods and services, V is the velocity of circulation, and M is the money supply in the economy. From this equation, we can see that money supply and price level have direct relationship. We should note that TQM assumes that V and T are constant in the short term, leaving only M and P variable. Consequently, when the money supply doubles, the price level in the economy also doubles. Thus, Central Banks can either increase or decrease the money supply in order to do the same in inflation. In the statement being analyzed, Central Banks are able to avoid wage-price spirals (which are considered P in the QTM) by pursuing a contractionary monetary policy. According to Mishkin (2004), lowering the money supply is done by raising discount rates which discourages bank borrowings, open market sale which tightens reserves and monetary base, and raise the reserve requirement among banks which shrinks the available funds for banks to grant as loans to borrowers. Also, another method usually done in open economies and has replaced monetary targeting is called inflation targeting. Inflation targeting is "an economic policy in which the central bank of a country estimates and makes public a projected or 'target' inflation rate and then attempts to steer actual inflation towards the target through the use of interest rate changes and other monetary tools (Inflation Targeting 2006)." Instead of directly controlling inflation by changing the level of money supply, central banks opted to manipulate interest rates. As interest rates and inflation are inversely related, the central bank raises interest rates if inflation appears to move above its target. Meanwhile, if inflation appears to be below the target, the central bank will lower interest rates. This policy has been adopted first by New Zealand in 1989. Inflation targeting has also been adopted by countries like the United States, Britain, South Korea, and Brazil. Inflation targeters have also set a time horizon over which to reach their targets. This usually depends on how high the starting rate of inflation is relative to the desired rate. Since, inflation targeting requires transparency; central banks periodically release inflation reports, and press statements (IMF 2003). 2. Outline the effects of such monetary policy on price expectations in the central bank's domestic economy. Inflation targeting, in order to be fully effective in curbing hyperinflationary expectations require transparency which

Evaluating a Health Promotion Website Essay Example for Free

Evaluating a Health Promotion Website Essay Health Information Technology (HIT) has been introduced into the National Health Service (NHS) in order to improve the quality, efficiency, safety and cost effectiveness of the delivery of health care. The application of computerized information technology in health care settings has so far played a vital role in improving the accessibility of information and has replaced more labour intensive and unproductive methods (Shekelle and Goldzweig, 2009). The Nursing and Midwifery Council (NMC, 2009) maintain that that nurses’ skills, knowledge and practice must be guided by the best available evidence. Often this evidence is found on the internet however many sources of information can be inaccurate so a good standard of critical evaluation is required (Kim, Eng, Deering et al 1998). To evaluate the quality and reliability of a chosen website the use of a framework can be helpful such as Roberts (2012) 5 C’s website tool. This framework evaluates five areas: credibility, currency, content, construction and clarity. The website The Royal College of Psychiatrists (RC Psych, 2012 a) was chosen to be evaluated because as a mental health nursing student this will develop an awareness of how information technology is implemented and applied to practice in health care. Additionally, by using Roberts (2012) 5 C’s framework to critically evaluate this website, knowledge and skills of the quality and reliability of HIT will be acquired. Main Body RC Psych is the professional and educational organisation for psychiatrist in the UK as well as being a registered charity (RC Psych, 2012 b). Their website is aimed at improving the lives of individuals affected by mental illness through educating the public. They claim to be at the head of developing and promoting best practice in mental health services through their education, training and research projects. Additionally they are involved in the publishing of the following world-class journals; British Journal of Psychiatry, The Psychiatrist, Advances in Psychiatric Treatment and International Psychiatry (RC Psych, 2012 c). Credibility Having gained a royal charter this shows us that the organisation has been recognised by the Queen and seconded by the government to call itself a registered organisation of the highest regard. As stated by the Privy Council Office (2012) organisations granted a Royal Charter must have a solid record of achievement. 75% of its members should be qualified to at least first degree level and the work completed by the organisation must be in the interest of the public. Professor Peter Tyrer is named as the Editor, of the website, and is said to be responsible for the editorial and production aspects of its publications in addition to the production of their online continuing professional development (CPD) e-learning resources and its sales and marketing (RC Psych, 2012 d). After researching Peter Tyrer it is evident that he is highly qualified within the field of mental health and is a professor of community psychiatry for the Department of Medicine, within the Imperial College London (Imperial College London, 2012). The website offers a vast amount of information around mental health including conditions, diagnoses, treatments and types of therapies. The advice provided is produced in the form of online leaflets for the use of the public as well as professionals. Within the website there is no evidence to show that the same specific authors are used regularly for their published articles however, at the end of each article the producer, editor and sometimes an author are named in addition to any expert that has been involved in the making of it. All these leaflets are produced by the RC Psych Public Education Editorial Board that is responsible for producing 300 educational leaflets. These leaflets have been accredited by the NHS Information Standard and subsequently gained numerous awards (Byrne, 2011). They have achieved Plain English and BMA patient information awards and have received consistently positive feedback for the web versions. The editor of these leaflets is Dr Phillip Timms who is currently employed as a consultant Psychiatrist for the South London Maudsley NHS Foundation Trust as well as a senior lecturer in psychiatry at King’s College, London (BMJ Masterclasses, 2012). This site’s web address shows that it is delivered by a UK based academic body as it has an ac.uk URL. This confirms their credibility and reassures the reader that they are qualified experts able to give advice in this field (Roberts, 2010). Combining all this it is apparent that the producers of the website have appropriate qualifications and expertise to offer advice in regards to mental health and supports the credibility of the organisation and the website. Some of these points also link into the other five C’s for example, the awards that the leaflets have received show that the content and clarity of their work has been assessed and found to be of an acceptable standard. Currency When using a website for personal development reasons or to recommend to a patient the information being accessed should be the most current evidence available that has been proven in practice. In regards to the advice provided on the website RC Psych (2012 e) states that they endeavour to update it every two to three years. This is reflected in the articles by showing the date the information was last updated and the date it will be reviewed in the future. Additionally RC Psych (2012 d) state that their information derives from the best evidence available at the time of writing and is updated regularly to reflect any changes in knowledge. Another way to determine the currency of information is to consider the references used to back it up (Roberts, 2010). If the sources are dated then what you are reading may not come from the most current research so there may be more up to date evidence elsewhere. The dates of the references used throughout the website are quite varied however, they do contain many recently published articles and up to date clinical guidelines so this suggests that they are committed to producing up to date information and evidence. This is seen in a leaflet on depression where RC Psyche (2012 f) cites a recently updated guideline by the National Institute for Clinical Excellence (NICE, 2009) which looks at the treatment and management of depression in adults. The currency and credibility of the website can also be reflected in its content and therefore this should also be assessed during the evaluation process. Content As well as being credible and current the content must be assessed for its objectivity, accuracy and completeness (Roberts, 2010). The content of a website can be judged by assessing what the website aims to achieve. This could be selling a product, persuading the audience to believe in something or to provide the reader with unbiased and up to date information on a specific topic (Roberts, 2010). RC Psych (2012 a) profess that they aim to improve the lives of individuals with a mental health illness. They aim to do this by educating people through making information and advice available on their website. This would indicate that it falls under the category of providing unbiased and up to date information to the public. The articles on the website are written in the third person so prevents the author from expressing personal opinion. Furthermore the leaflets produced provide links to various other sources of information on the topic being discussed. This encourages further reading which provides a complete and balanced view preventing bias and, where treatment is recommended, allows the reader to make an informed choice (Roberts, 2010). The references used within the website are taken from well known and respected sources that provide current and acknowledged recommendations such as NICE, the Department of Health (DOH) and various field related journals. This indicates that the website seeks to provide and maintain up to date, accurate and unbiased information that concurs with information available in books and journals on the same subjects. Construction The construction of a website, for example, the layout, colours, fonts, sizes and ease of access can determine whether or not a reader believes the website is of good quality or not (Roberts, 2010). A study by Lindgaard, Fernandes and Dudek et al. (2006) found that the first 50 milliseconds that the reader is exposed to a website can cause the reader to form an opinion as to whether a website is worth using or not. On first impression, the website in question looks colourful and professional. The eye is drawn to the bolder writing that states the websites purpose and there are minimal graphics to distract the reader from this. There is a lot of information on the home page of the website and could be considered cluttered, however it is divided into clear sections and appears well organised, giving clear subheadings which guide the reader to their areas of interest easily. The main colour used for the website is grey and presents a professional look. It is subtle but effective in breaking up different sections of the website without being obtrusive. The font size is varied throughout but mainly of a larger size making the website easy to read. There is no option for the reader to change the font sizing or colour which could cause some difficulty as it is not possible to suit everyone with one font size, style and colour (Roberts, 2010). As this is a registered charitable organisation it would be unfair to expect them not to promote money making offers. There are two links to buy books that are published by the royal college of psychiatrists and an option to support them in future development projects. They have presented these links as the last things you would come across on the website. Because of where they are situated it comes across to the reader that they are more interested in providing free and accessible information to educate the pubic over making money. Clarity Clarity is another important area to consider when evaluating a website. Due to our continually growing multi-cultural society it is important that websites cater for all groups in society and not just the English readers. One part of the website caters for professionals working in psychiatry and another to the public wanting to learn more about mental illness. Both sections are presented and written in a way that is understandable to the general public. In the public section they offer their advice in 21 different languages in addition to visual aids using BSL sign language as well as audio pod casts and printable versions. This shows that they have made an effort to cater for diverse cultures and needs. Unfortunately it appears that they have not taken into account the needs of people with dyslexia. Approximately two million people in the UK population are affected by dyslexia of which around 35 to 40 percent experience visual disturbances when reading (Dyslexia Action, 2012). The British Dyslexia Association, (2012) states that the reading ability of an individual with dyslexia can be negatively affected by bright white backgrounds and the use of too much text. By changing the background to an off white colour and spreading out the information over larger areas this would cater for yet another group in society. Conclusion Having the ability to access health information via the internet has given nurses the capacity to constantly improve their knowledge base and skills knowing that they are delivering the best possible care derived from the best available evidence. The RC Psych website not only offers nurses a place to go and build on their knowledge and skills but it also provides them with a safe, reliable and easily accessible knowledge base that they can confidently refer their patients to. The evaluation of this website has shown that the website contains quality, reliable evidence and could be recommended to anyone interested in the field of mental illness. In addition to this the use of the five C’s evaluation tool has provided a great foundation to the development of evaluating skills. It has given an opportunity to discover that the evidence being accessed is suitable for developing personal practice and to help educate patients and is accepted within the NMC code of conduct (Roberts, 2 010).

Language, Culture And Thought

Language, Culture And Thought ABSTRACT Language is the basic means of communication in every culture. Language expresses cultural reality of the speakers in every society. Language is a socially acquired phenomenon and all the culturally bound languages differ dramatically from one another in terms of the description of the natural world. Language not only embodies cultural reality but it is also viewed as a symbol of social identity This assignment explores different ways in which the world is described by the speakers of different languages due to their cultural diversity and its effect on different ways of thinking about the world. The study reviews the description of spatial terms, temporal terms, substances and objects by different speakers of different languages belonging to different cultures and its effect on the ways of thinking about the world around us. Keywords: language, thought, culture, space, time, objects, substance, linguistic relativity, Whorf, Sapir LANGUAGE, CULTURE AND THOUGHT Language is a complex phenomenon for human communication. In the domain of communication each language of the world is different from the other on both lingual and cultural grounds. This diversity in world languages from linguistic standpoints ranges from the clear distinction of pronunciation and vocabulary to the more complex differences of grammar. The diversity of cultural standpoints related to different languages can be seen in the variety of cultural connotations assigned to these linguistic codes. The representation adopted by different languages to the same sentence such as, the professor delivered the lecture, is different to each culture. In English, the verb delivered marks the past tense. In Indonesian and Mandarin, the verb never changes to mark tense. In Russian, the verb is changed to mark tenses and gender, so if the subject is female instead of male, then a different verb is used from the verb that is used for the male. Similarly, in Russian, the verb is also change d to mark whether the whole lecture was delivered or only part of it was delivered by the professor. In Turkish, the verb even specifies weather the delivery of lecture was witnessed or it was a gossip. It is apparent the speakers of different languages see the different aspects of the world to use their language properly (Gumperz and Levinson: 1996) Do these features of language affect the way their speakers think about the world? The idea that thought is shaped by language is most commonly associated with the works of American linguistics Edward Sapir and Benjamin Lee Whorf. According to Whorf, the categories of every human language present a way of perceiving, analyzing and acting in the world. The strong view of the Sapir Whorf Hypothesis that thought and actions are completely determined by language has been rejected in the field of linguistics. However, answering a less deterministic weaker version does language affect thought has proven to be a very difficult task for the researchers. In recent times researches have found new evidences which highlight the effects of language on thought. This assignment discusses the effect of language on peoples thinking of space time, substance, and objects. Research Question To what extent are the world views and mental activities of members of a social group shaped by, or dependent on, the language they use? LANGUAGE AND CULTURE The word culture is etymologically derived from the Latin word cultura, which means to cultivate and different philosophers such as Voltaire, Hegel, Humboldt and Kant assign different meaning to it. In antiquity, culture was referred to the opposite of nature, something that is willingly produced by man. (Dahl: 2001) In 18th century, the term was used to describe elite and high-culture concepts in continental Europe. According to Dahl, culture is a collectively held set of attributes, which is dynamic and changing over time. It is the totality of values, beliefs, basic assumptions as well as code of social behavior in a society. Individual are made by culture on one level and culture is made by individuals on the other. (Dahl: 2001) Language is the basic means for communication in a society. Language is primarily related with culture in the domain of communication. Speakers communicate their ideas, feelings and facts with language both in inter and intra societal setups. Hence, we can say that language expresses cultural reality. (Kramsch, 3: 2005) Language is not only a mean of expressing experience but it also creates experience. Speakers give meaning to such an experience through various mediums, such as, mass media, internet, telephonic conversations, presentations, etc. The medium with which language is used also designate meaning to experience, for example, the phonological and supra-segmented phonological features of language. Hence, we can say that language embodies a cultural reality. (Kramsch, 3: 2005) In every society the collective set of social values, beliefs, norms and mores are transmitted from one generation to another by the process of socialization. Hence, culture is a socially acquired knowledge. If we observe language from this standpoint, it is observable that language is also transmitted from one generation to another through the process of socialization. It is, therefore, evident that language is also a socially acquired knowledge. (Horton and Hunt: 2005) Finally, the speakers of one language identify themselves with a particular group through their peculiar way of using language. This characteristic of individuals in a group signify that language is viewed a symbol of social identity. Hence, we can say that language symbolized culture. (Kramsch, 3: 2005) EFFECTS OF LANGUAGE ON THOUGHT Space Languages differ from each other in the description of the setting of space. For example, In English, it is different to put something in which it is contained (the orange in the plate, the letter in envelope) and putting things onto surfaces (the orange on the table, the sticker on refrigerator door). `In Korean, putting an orange in a bowl uses a different term (nehta) from putting a letter in an envelope (kitta). Additionally, putting a letter in envelops and putting sticker on the refrigerator is both expressed by the similar term Kitta because both involves close fitting. McDonough et al. (2000) conducted a research to test whether the linguistic differences between English and Korean speaker affects the representation of spatial relations. He showed scenes containing examples of tight and loose fit to Korean and English adults. The Korean adults looked longer at the example scenes compared to English adults. After that they were shown tight fit on one screen and loose fit on another and were asked to distinguish between them. The Korean adults were quick to identify between the tight fit and lose fit, whereas, the English adults were unable to distinguish between the tight fit and the lose fit, and they took relatively longer time than the Korean adults. Further, when these speakers were given several example of tight fight and one example of lose fight (or vice versa). The Korean adults could easily pick out the odd ones, but the English adults could not. This study concluded that the distinctions of settings reinforced by the particular languages r emained central in the representation of their spatial settings. Time Languages are also different from one another because of the difference in the descriptions of time. In English language, the conception of time is represented in horizontal terms. For example, students may ask the teacher to move the test forward, or the teacher may push the test back for the students. Similarly, the speakers of Mandarin language also use terms qian (front) and hou (back) to represent time in horizontal terms. However, the speakers of Mandarin also uses vertical terms such as shang (up) and xia (down) to describe time which in English signifies last and next respecitively. The use of the vertical terms of time in Mandarin is more common in than the horizontal terms of time. In a study conducted by Borodistsky (2001), it was found that the Mandarin speakers think about time vertically even when they are thinking for English. The Mandarin speakers were quick to answer questions such as March comes before April when they were shown objects in a vertical order before questioning compared to the English speakers. Similarly, the English speakers gave quick response to this answer when they were showed objects arranged in a horizontal order before the question were asked from them for the study, compared to the Mandarin speakers. The researcher in another attempt gave English and Hebrew speaking adults sets of pictures that showed some kind of temporal progression such as, pictures of a man aging, or a plant growing, or an apple being eaten. (Boroditsky: 2002) They were asked to arrange the shuffled photos on the ground to show the correct temporal order. They were tested in two separate sittings, each time facing a different direction. The English speakers arranged the cards in a way that time progresses from left to right. The Hebrew speakers arranged the cards in a way that time progresses from right to left, showing that the direction of writing plays a role in the progression of time. Objects Language also differs in the way the names of the objects are arranged into grammatical categories. Unlike English, many languages assign gender to all the nouns. A recent set of studies suggested that grammatical gender given to objects by the language influence the mental representation of these objects by the people (Boroditsky and Schmitt: 2003). The researcher gave pictures of people (male and female) and also pictures of objects (having opposite genders in German and Spanish) to two groups of German and Spanish speakers. They were asked to rate similarities between the pictures of objects and pictures of people. Both groups rated grammatically feminine objects to be more similar to females and grammatically masculine objects more similar to males as described by their languages. This test was a complete nonlinguistic test. In Spanish, the word for key is feminine, while the German word for key is masculine. The Spanish and German speakers were asked to provide descriptive adjectives for different objects. German speakers described keys as hard, heavy, jagged, metal, and useful, while Spanish speakers described them as golden, intricate, little, lovely, and shiny. For the word bridge, which is feminine in German and masculine in Spanish, the opposite happened. Germans described bridges as beautiful, elegant, fragile, and peaceful, while Spanish speakers said they were big, dangerous, long, and strong. CONCLUSION Languages appear to influence many aspects of human cognition: evidence regarding space, time, objects, has been reviewed in this assignment. Further studies have also found effects of language on peoples understanding of numbers, colors, and shapes, events, and other minds. Considering the many ways in which languages differ, the findings reviewed here suggest that the mental lives of people who speak different languages may differ much more than previously thought Beyond showing that speakers of different languages think differently, these results suggest that linguistic processes are pervasive in most fundamental domains of thought.

Compensation And Benefit Strategies For Current Generation Workforce Business Essay

Compensation And Benefit Strategies For Current Generation Workforce Business Essay Compensation and benefits is one of the many functions of human resource management. Compensation plans have not always been considered a strategic business initiative. However, with the effects they have on recruiting, retaining and motivating people compensation design is a critical element to achieving success. Introduction: Employees are the most valuable assets of any organization. The main purpose of human resource management (HRM) is to manage the development and performance of people employed in an organization. Some of the important tasks of HRM include attracting and retaining employees with the right skills and abilities, match people to the right positions within an organization and to align employees goals and objectives with that of the organizations. All these key tasks are directly or indirectly affected by compensation and benefits plan which the human resource management decides. Compensation and benefits redefines value and success in any workplace. An efficient compensation and benefits strategy is seen as an opportunity by organisations to differentiate themself from their competitors, who may otherwise take away talented workforce. Hence, organisations need to know the evolving needs of the working population to make sure they offer the most competitive package. This paper adopts a general approach which may be relevant to number of different organizations or industries and discusses how the new generation workforce is making an impact and forcing organizations to change their traditional compensation strategies. This paper also examines how companies located in different sectors and countries look at compensation and benefits aspect of human resource management, to attract and retain their highly qualified staff. Finally the paper summarizes and analyses few articles related to HR practices in compensation and benefits, and see if they fit into organizations seeking best practice. Compensation and benefit strategies for current generation workforce: Compensation is a major driver in the success of any organization and it is also true that it is one of the highest expenses for an organization. Therefore, compensation is always under heavy scrutiny by the top executives of a company. Compensation is not only carefully analysed by the management, but also analysed by employees and prospective employees. If properly executed, effective compensation design can improve organizational effectiveness, support human capital requirements of the organization, and motivate the employees to achieve key corporate strategic and financial goals (Ellig, B. R. 2007). Effective compensation and benefits strategy is no easy task, especially with the large organizations and ever-changing demands of the workforce. To construct the most attractive compensation and benefits package, organizations must now tackle the fundamental landscape change in workforce requirements and demands. Employees born in the 60s and 70s (generation X) have ceded place to a new generation of workers (generation Y). A flexible approach to compensation and benefits is now required to satisfy the different priorities of these generations. Planning for the future has long been the motto governing Generation Xs approach to their careers and so, traditionally, generous pension provision has been the key to attracting workers. But for Generation Y, priorities have changed. For instance, in a research conducted last year, just 4% of 16-24 year olds were attracted to their employer because of their pension contribution, compared to 17% of 45-54 year olds (HR magazine, March 2011). Certainly, younger workers are increasingly questioning the customs and traditions of the workplace and now pension plans, previously synonymous wi th security, are considered unstable. In present day environment, employees have become much more educated on the possibilities available to them with sources like online salary calculators, industry chat rooms and so on (Schneider, B., Paul, K. B. 2011). Employees now know more about current market pay levels, new opportunities and how to successfully negotiate compensation packages. Keeping these changes in mind organizations include many other noncash components to the potential cash components to come up with a fair, competitive and an attractive compensation program (Schneider, B., Paul, K. B. 2011). Employee benefits are indirect forms of compensation provided by the organizations to their workforce as part of an employment relationship. To compete for quality employees in todays marketplace, employers must do more than offer a fair salary. Employees also look forward for a good benefits package; in fact employees have grown accustomed to generous benefits programs, and join an organization expecting them. Employee benefits exist in organizations all around the world and the levels and range of these benefits vary between countries. Some of the benefits companies offer can be like company accommodation, company provided vehicles, health insurance, retirement benefits, easy loans and so on. Compensation and Benefit strategies used in organisations: As per Kevin OConnell, there are three primary components in a strategic compensation plan. First, building a solid foundation for the plan to understand the business strategy is required. Where the company is now and where is it heading to in the future, what are its short term and long term goals. This will provide the necessary information needed to construct a solid framework for the design of a compensation plan. Second, understanding organizational capabilities and how the role of compensation plays a key role to achieve organizational goals is critical. This will provide an excellent understanding of what positions are the most critical to a companys future success and the various compensation options available to motivate employees to achieve this success. Third, aligning compensation with human capital management initiatives will help develop a plan that allows the company to successfully recruit and retain employees (OConnell, K. 2007). We will see how different organizations have overcome significant business challenges by re-evaluating and strategically designing new compensation plans. By finding the right balance between organizational goals and employee needs, compensation plans can play a strategic role in the success of a business. Jamba Juice case: Jamba juice was founded in 1990. It is a leader in blended-to-order fruit smoothies and fruit juices. (Source: http://www.jambajuice.com/) Since its inception, it found employee retention as a major problem. It was located in San Francisco bay area and therefore the employees had more options with other employers situated there. A lot of these employers were technology based and offered more generous financial benefits than the food retailer. To tackle this problem, Jamba Juice introduced a compensation plan, called J.U.I.C.E plan which allows the general manager to receive a percentage of stores cash flow depending on their businesss final performance. It allows manages a share in the profits over a period of three years. When the general managers increase the year-to-year sales, money accrues in a retention account, which is payable only in three year cycles. This retention account not only provides short term retention benefits, but also incentives to stay with Jamba. Also, execut ives at managerial positions are offered stock options. And, when assistant managers are promoted, their general managers also receive a cash award of $ 1000 for their development efforts. In a highly competitive industry Jamba was able to successfully reduce turnover. (Mellow Jeffrey, 2011) The MAERSK way: MAERSK not only rewards absolute performance but also focuses significantly on performance relative to peers, says Alex Penvern, Global Head of Group Compensation, Rewards and Executive HR. This had not always been the case. Less than five years ago, rewards in the company were characterised by confidential bonuses, awarded with very little transparency. One of Penverns first challenges when he joined MAERSK in 2008 was to create a scalable, measurable and quantifiable compensation and benefits structure that was understandable and could, over time, be rolled out uniformly across the organisation (Source: http://www.maersk.com/Pages/default.aspx). He spoke to CEOs of each business within the group and focused on the executive compensation structure that focussed on relative distribution. During these conversations he discussed range of different performance criteria, happening in annual sessions which were a part of the performance management cycle. The outcome was a relative performance distribution of the companys most successful and less effective performers. The company believes that our people are motivated by this constant striving to do even better, says Penvern. You can never rest on your laurels or spend too long patting yourself on the back, because you know how hard everyone else is running. We want people who thrive in this atmosphere. This is brought in practice and reinforced by a carefully considered distribution of rewards to the highest performers. Since the introduction of the pay-for-performance scheme, fewer very high performers are securing a significantly larger share of the bonus on offer. The highest performers receive nearly double the bonus opportunity that they did few years back but to earn that bonus they need to keep up or stay ahead of their peers or market. While Penvern discusses the value the company creates in this performance culture, he believes that the transparency of its bonus system is just as important. The lesson is clear. Company has to build employee engagement and drive performance both by having a clear and transparent compensation scheme that links pay and performance and by communicating this consistently in order to reinforce and reinstate the belief that pay and performance are linked. Design and Engineering group in Delaware: A small to medium design and engineering group in Delaware, employs engineers and designers, most of them who are in their 30s. Most of the employees were the primary wage earners in their families and have several dependents. This company pays wages that are slightly lower than those of its competitors, but it offers a fully paid insurance program that includes health, pharmacy, dental and eye care coverage for employees and their dependents. By meeting the employees benefit needs the company enjoys very low employee turnover (McConnell, J. 2003 Pg. 235). Mississippi garment manufacturer: A garment manufacturer in Mississippi pays on a piecework basis. Most piecework plans in garments factory in general must guarantee pay equal to at least the legal minimum wage. In this specific company, new employees receive an hourly wage until they have acquired the skills required to meet performance standards. Then the manufacturer pays them on the basis of the number of pieces they produce in an hour. The company has designed its compensation approach to retain good employees i.e. those who are good performers (McConnell, J. 2003 Pg. 232). Conclusion: Compensation whether in the form of salary or benefits is a key element in both attracting and retaining employees. With careful and logical thinking companies should design a program that will significantly contribute to obtaining and retaining the type of employers who perform best for that company. Compensation professionals play a critical role in enabling organizations to execute their business strategy. The examples in this paper demonstrate how organizations have overcome significant business challenges by redesigning their compensation plans as seen in Jamba and MAERSK. We also have seen how a small company like the garment manufacturer in Mississippi use compensation package to retain highly performing people and do away with low performing people. Some companies like the one in Delaware uses non monetary benefits to keep their workforce happy and content. Compensation and benefits have a negative effect on the company if not properly managed. Collapses of big companies like Enron and Global crossing are examples. These are direct results of managerial behaviour aimed at short term profit maximization. That behaviour is in turn a result of the executive pay structure that measure and reward performance based short-term financial results. Compensation experts need to be at the focal point where strategy, organizational effectiveness and human capital management congregate. They need to have a clear understanding of the business and organizational issues as well as the future direction of the company. Strategic compensation and benefits design is not a sound bite to impress management. Its the process behind compensation plan design that links and binds strategy, organization effectiveness and human capital together. By balancing organizational, employee and business needs with a companys financial and strategic goals it is possible to develop the right compensation strategy to motivate, reward and sustain high levels of performance. By finding this balance, a company can effectively use compensation to execute and achieve desired business results.