Hierarchical Bayesian Aldrich-McKelvey Scaling in R via Stan

Example Data

For illustration, we will use data from the 1980 American National Election Study (ANES). This is the same data set that serves to illustrate the original AM model in the basicspace package. The data set is included in the hbamr package and can be loaded by running data(LC1980).

The data set contains respondents’ placements of themselves and six stimuli on 7-point Liberal-Conservative scales. The stimuli in question are: The Democratic and Republican parties, Democratic presidential candidate Jimmy Carter, Republican candidate Ronald Reagan, independent candidate (and former Republican) John B. Anderson, and Ted Kennedy (who challenged the incumbent Carter, but failed to win the Democratic nomination).

We load the data and re-code missing values as follows:

library("hbamr")
data(LC1980)
LC1980[LC1980 == 0 | LC1980 == 8 | LC1980 == 9] <- NA 
self <- LC1980[, 1]
stimuli <- LC1980[, -1]

head(stimuli) 

##    Carter Reagan Kennedy Anderson Republicans Democrats
## 1       2      6       1        7           5         5
## 8       4      6       4        7           6         4
## 9       3      6       3        3           6         2
## 10      6      4       3        3           5         4
## 11      7      2       5        5           7         5
## 13      6      6       2        5           7         4

Preparing the Data

The function prep_data() serves to prepare the data. This function can be run ahead of fitting the models, or it can be run implicitly as part of a single function call to fit the models (as shown below). The function takes a vector of \(N\) ideological self-placements and an \(N \times J\) matrix of stimulus placements. It applies a set of inclusion criteria, performs any necessary data transformation, and returns a list of data suited for sampling in rstan. The stimuli data are stored in a vector as a long-form sparse matrix. If the stimuli data include column-names, these will be preserved for later use.

Any missing data must be set to NA before use. The prep_data() function allows the user to decide how many missing values should be permitted per respondent by specifying the argument allow_miss. (The default is allow_miss = 2. Alternatively, the argument req_valid specifies how many valid observations to require per respondent. The default is req_valid = J - allow_miss, but, if specified, req_valid takes precedence.) Similarly, the user may specify how may unique positions on the ideological scale each respondent is required to have used when placing the stimuli in order to be included in the analysis. The default is req_unique = 2, which means that respondents who place all stimuli in exactly the same place will not be included.

The data provided to prep_data() can be centered, but they do not have to be: The function will detect un-centered data and attempt to center these automatically, assuming that the highest and lowest observed values in the data mark the extremes of the scale.

To use the prep_data() function on the example data using the default settings, we would run:

dat <- prep_data(self, stimuli)

Users who want to keep other covariates for subsequent analysis, may find it useful to run prep_data() separately from the call to fit the models. The list returned by this function includes the logical vector keep, which identifies the rows in the original data that have been kept. If we had a data set x containing covariates, and used prep_data() to produce the list dat, then we could use x[dat$keep, ] to get a subset of x corresponding to the data used in the analysis. (The order of the individuals/rows in the data remains unchanged by the functions in this package.)

Models

This package provides several alternative models that can be selected using the names below. Users who are unsure which model to use are advised to use the default HBAM model. If speed or sampling diagnostics are an issue, HBAM_MINI may provide a useful alternative.

HBAM is the default model, which allows for scale flipping and employs hierarchical priors on the shift and stretch parameters. It also models heteroskedastic errors that vary by both individual and stimuli. Compared to the model in Bølstad (2023), this version has been slightly revised to provide faster sampling. A key difference from the original model is that the respondent positions are not treated as parameters, but rather calculated as a function of the self-placements and the individual-level parameters. This makes the model considerably faster, while yielding very similar results.

HBAM_MULTI is a version that models differences between groups defined by the user. It requires an integer vector identifying the groups to be supplied as the argument group_id. The model gives each group separate hyperparameters for the locations of the prior distributions for the shift and stretch parameters. Rather than shrinking the estimates toward the mode for the whole dataset, this model shrinks the estimates toward the mode for the group. The vectors of hyperparameters are called mu_alpha and mu_beta and are constructed to have means of 0. The scales of the priors on these hyperparameters can be set by the user via the arguments sigma_mu_alpha and sigma_mu_beta. The default values are B / 5 and .3, respectively. (Here, B measures the length of the survey scale as the number of possible placements on one side of the center.) One potential use for this model is to supply self-placements (appropriately transformed) as group_id, and thus give each self-placement group its own prior distribution for the shift and stretch parameters.

HBAM_NF (formerly HBAM_0) is a version of the HBAM model that does not allow for scale flipping. This may be useful if there are truly zero cases of scale flipping in the data. Such scenarios can be created artificially, but may also arise in real data. For example, expert surveys appear unlikely to contain many instances of scale flipping. For data that contain zero cases of flipping, models that allow for flipping contain superfluous parameters that lead to inefficient sampling. Models that do not allow for flipping will sample faster and typically yield slightly more accurate estimates. Such models are therefore preferable when no flipping is present.

HBAM_MULTI_NF is a version of the HBAM_MULTI model that does not allow for scale flipping.

HBAM_MINI is a version of the HBAM model that assumes the prediction errors in the stimuli placements to be homoskedastic. This model tends to sample faster faster than the standard HBAM model while yielding very similar point estimates. For large datasets, this model may provide a reasonable compromise between model complexity and estimation speed.

FBAM_MINI is a version of the HBAM_MINI model with fixed hyperparameters to allow fitting via optimization rather than MCMC – which can be useful for large data sets. This model allows the user to specify the scales of the priors for the shift and (logged) stretch parameters via the arguments sigma_alpha and sigma_beta. The default values are B / 4 and .35, respectively. These defaults are intended to be realistic and weakly informative. As with the other models, the default values of the scale-dependent priors are automatically adjusted to the length of the survey scale. Users who want to control the degree of shrinkage of the individual-level parameters may find it useful to fit this model (or other FBAM models) via either MCMC or optimization.

FBAM_MULTI is a version of the FBAM_MINI model that shares the group-modeling features of the HBAM_MULTI model. It allows the user to set the scales of the priors for the shift and stretch parameters via the arguments sigma_alpha and sigma_beta, and set the scales of the priors on mu_alpha and mu_beta via the arguments sigma_mu_alpha and sigma_mu_beta.

FBAM_MULTI_NF is a version of the FBAM_MULTI model that does not allow for scale flipping.

HBAM_R_MINI is a version of the HBAM_MINI model that incorporates the rationalization component of the ISR model in Bølstad (2020). This model requires additional data to be supplied as the argument pref: An N × J matrix of stimuli ratings from the respondents. The rationalization part of the model is simplified relative to the original ISR model: The direction in which respondents move disfavored stimuli is estimated as a common expectation for each possible self-placement on the scale.

BAM is an unpooled model with wide uniform priors on the shift and stretch parameters. It is similar to the JAGS version introduced by Hare et al. (2015). This model is mainly provided to offer a baseline for model comparisons. While it is simple and fast, this model tends to overfit the data and produce invalid posterior distributions for some respondent positions (Bølstad 2023).

HBAM_MAX treats the latent respondent positions as parameters – in contrast to all other models in this package. The advantage of doing so is that the uncertainty of the respondent positions may be estimated more accurately. A key disadvantage is the model takes longer to fit. Another potential disadvantage is that the hierarchical prior on the respondent positions shrinks them toward their common mean, which may be undesirable if the distances between respondents and stimuli are of interest.

HBAM_2 is deprecated and replaced by the more general HBAM_MULTI model.

These models can also be used in situations where self-placements are not available and the only goal is to estimate stimulus positions. This can be achieved by supplying a vector of zeros (or random data) instead of real self-placements: self = rep(0, nrow(stimuli)).

To see the Stan code for any of the models, use the show_code() function:

show_code("HBAM_MULTI")

Summary of Model Features

Fitting

The hbam() function can be used to fit all models in this package and it returns a stanfit object. The default model is HBAM, while other models can be specified via the argument model. For each model, hbam() selects a suitable function to produce initial values for the sampling algorithm.

Unless the user supplies pre-prepared data via the data-argument, hbam() will implicitly run prep_data(). It therefore takes the same arguments as the prep_data() function (i.e. self, stimuli, allow_miss, req_valid, and req_unique).

To fit the HBAM model using the default settings, we would run:

fit_hbam <- hbam(self, stimuli)

To fit the HBAM_MINI model while requiring complete data for all respondents, we would run:

fit_hbam_mini <- hbam(self, stimuli, model = "HBAM_MINI", allow_miss = 0)

If we wanted to run the prep_data() function separately before fitting the model, we would supply output from prep_data() as the argument data:

dat <- prep_data(self, stimuli, allow_miss = 0) 
fit_hbam <- hbam(data = dat)

To fit the HBAM_MULTI or FBAM_MULTI model, we would need to supply a vector identifying the groups of interest. One option would be to give each self-placement group their own prior distributions for the shift and stretch parameters. (Note that supplying the raw self-placements as group_id works in this case because the self-placements are stored as positive integers from 1 to 7 – otherwise they would have to be re-coded.) If we decided to fit the FBAM_MULTI model, we could also specify our own scales for the priors on key parameters:

fit_fbam_multi <- hbam(self, stimuli, model = "FBAM_MULTI", group_id = self, 
                       sigma_alpha = .8, sigma_mu_alpha = .5, 
                       sigma_beta = .4, sigma_mu_beta = .25)

hbam() uses rstan::sampling(), and any additional arguments to hbam() will be passed on to the sampling function. By default, hbam() will run 4 chains and detect the number of available CPU cores. If possible, hbam() will use as many cores as there are chains. The other sampling defaults are: warmup = 1000 and iter = 2000. These settings can all be overridden in the hbam() call.

Fitting via Optimization

For very large data sets, optimization can be a useful and much faster alternative to MCMC. However, the optimization feature provided in this package only provides maximum a posteriori (MAP) point estimates.

The fbam() function fits FBAM models using rstan::optimizing(). The fbam() function works just like the hbam() function, except the arguments for rstan::sampling() do not apply. To fit the FBAM_MINI model using default settings, we would run:

fit_fbam <- fbam(self, stimuli)

Execution Times

Models of the kind included in the hbamr package face an inevitable trade-off between nuance (i.e. model complexity) and execution times. For small data sets (like ANES 1980), all models provide reasonable running times, but for large data sets (like ANES 2012), more complex models tend to get slow. In these cases, the HBAM_MINI model may be a useful alternative. For very large data sets, fitting the FBAM_MINI model via optimization may be the best alternative.

Plotting

The hbamr package uses ggplot2 for plotting (which means ggplot-themes can be added to the plots). The function plot_stimuli() plots the marginal posterior distributions of all stimuli in the data. By default, it will fill the distributions with shades from blue to red depending on the position on the scale. The argument rev_color = TRUE will reverse the order of the colors.

plot_stimuli(fit_hbam)

In this example, we see that John B. Anderson – the former Republican who ran as an independent candidate – gets a wider posterior distribution, suggesting that voters were more uncertain about where to place him relative to the others.

The function plot_respondents() plots the distribution of estimated respondent positions. It illustrates the uncertainty of the estimates by calculating the population density for each of a set of posterior draws. The default is to use 15 draws for each respondent, but this can be altered by specifying the argument n_draws. The plot_respondents() function also plots the estimated stimulus positions by default, but this behavior can be turned off by adding the argument inc_stimuli = FALSE.

plot_respondents(fit_hbam, n_draws = 10)

Users who want to customize the plots further can obtain the underlying data by using the function get_plot_data(). This function accepts the same n_draws-argument as plot_respondents(). The output is a list of three tibbles: The first element contains the posterior mean stimulus positions, as well as the \(x\)- and \(y\)-values of the posterior modes (which can be useful for labeling the distributions). The second element contains the posterior draws for the stimulus positions (which can be used to calculate marginal posterior densities). The third element contains the selected number of posterior draws for each respondent (which form the key ingredient for the plot_respondents() function).

The function plot_over_self() plots the distributions of key parameter estimates over the respondents’ self-placements. In addition to a stanfit object produced by hbam(), this function requires the list of data produced by prep_data(). The function also takes the argument par, which can be either of the following: "alpha", "beta", "abs_beta", "lambda", "chi", and "eta". "abs_beta" calls for the absolute value of beta to be used. By default, the function uses posterior median estimates, but this can be changed by specifying estimate = "mean".

This function can, for instance, be used to assess the accuracy of respondents’ answers. In particular, when the argument par = "eta" is specified, the plotting function will display \(\sqrt{\eta_i} / J\), which equals the average error for each individual (the mean of \(\sigma_{ij}\) for each \(i\) across \(j\)). The point estimates will still be calculated using the posterior median, unless the argument estimate = "mean" is added. (This plot is not available for the homoskedastic variants of the model (such as HBAM_MINI) as these do not estimate individual \(\eta\) parameters.)

plot_over_self(fit_hbam, dat, "eta")

The plot_over_self() function may also be used to compare results from several models by supplying the model fits as a list. To compare the distributions of estimated shift parameters from the HBAM and HBAM_MINI models, we would run:

plot_over_self(list(fit_hbam, fit_hbam_mini), dat, "alpha")

For these models, the draws for \(\beta\) combine the separate parameters for each flipping-state. The absolute value of \(\beta\) may therefore be better suited for examining the extent to which each individual stretches the ideological space. To inspect the distribution of these values across self-placements, we would run:

plot_over_self(list(fit_hbam, fit_hbam_mini), dat, "abs_beta")

The pattern above, where respondents with more extreme self-placements have more extreme \(\beta\) parameters, is exactly the kind of differential item functioning that the models in this package are intended to correct for: These respondents tend to place both stimuli and themselves further out on the scale than others do, thus appearing more extreme in comparison.

It may also be useful to inspect the distribution of the scaled respondent positions over the self-placements. This illustrates the extent to which the model has transformed the original data. In this example, the impact of the models is generally modest, although a few respondents have been detected as having flipped the scale, and thus have had their self-placement flipped back.

plot_over_self(list(fit_hbam, fit_hbam_mini), dat, "chi")

Posterior Summaries

The package also contains a wrapper for rstan::summary() called get_est(). This function takes the arguments object – a stanfit object produced by hbam() or a list produced by fbam() – and par – the name of the parameter(s) to be summarized. The function returns a tibble, which by default contains the posterior mean, the 95% credible interval, the posterior median, the estimated number of effective draws, and the split R-hat. One can obtain other posterior quantiles by using the argument probs. To get a 50% credible interval (and no median), one would add the argument probs = c(0.25, 0.75). To include the Monte Carlo standard error and the posterior standard deviation, use the argument simplify = FALSE. (When applied to outputs from fbam(), get_est() only returns point estimates.)

The posterior draws for the stimulus positions can be summarized as follows:

get_est(fit_hbam, "theta")

## # A tibble: 6 × 6
##     mean `2.5%`  `50%` `97.5%` n_eff  Rhat
##    <dbl>  <dbl>  <dbl>   <dbl> <dbl> <dbl>
## 1 -1.02  -1.10  -1.02   -0.941 2160. 1.00 
## 2  1.62   1.55   1.62    1.68  2013. 1.00 
## 3 -1.68  -1.76  -1.68   -1.60  2473. 1.00 
## 4 -0.453 -0.574 -0.453  -0.333 3159. 1.00 
## 5  1.42   1.36   1.42    1.48  1912. 1.00 
## 6 -1.19  -1.25  -1.19   -1.13  2404. 0.999

The equivalent call for the respondent positions is:

get_est(fit_hbam, "chi")

## # A tibble: 643 × 6
##       mean `2.5%`    `50%` `97.5%` n_eff  Rhat
##      <dbl>  <dbl>    <dbl>   <dbl> <dbl> <dbl>
##  1  0.965  -1.96   1.27      2.62  2862.  1.00
##  2  1.21   -0.613  1.26      2.57  3421.  1.00
##  3  1.39   -0.121  1.44      2.35  3212.  1.00
##  4  0.0128 -1.80   0.00469   1.76  3745.  1.00
##  5 -0.197  -2.04  -0.253     1.89  4089.  1.00
##  6  1.45   -1.78   1.68      2.94  2336.  1.00
##  7 -0.157  -1.15  -0.159     0.857 2975.  1.00
##  8  0.943  -0.738  0.991     2.01  2924.  1.00
##  9  0.587  -0.857  0.623     1.73  3691.  1.00
## 10  0.125  -1.02   0.136     1.14  3654.  1.00
## # ℹ 633 more rows

Diagnostics

The rstan::sampling() function that hbam() uses automatically performs a number of key diagnostic checks after sampling and issues warnings when a potential issue is detected. The authors of rstan emphasize diagnostics and careful model development, and users of rstan will more frequently encounter warnings than users of rjags. One warning users of this package may encounter is that the Bulk or Tail Effective Sample Size (ESS) is too low (see https://mc-stan.org/misc/warnings.html). The most straightforward solution to this issue is to increase the number of posterior draws, using the iter argument. However, this increases the computational load, and users should consider carefully what level of accuracy they need.

Because the hbam() function returns a stanfit object, the model fit can be examined using the full range of diagnostic tools from the rstan package. Users should consult the rstan documentation for details on the various diagnostic tests and plots that are available. One example of the available tools is traceplot():

rstan::traceplot(fit_hbam, pars = "theta")

Cross-Validation

A useful way to compare alternative models is to estimate their out-of-sample prediction accuracy. More specifically, we can estimate their expected log pointwise predictive density for a new data set (ELPD). It is most straightforward to do so for the stimuli placements, and this is informative because these form the basis for estimating most individual level parameters. A similar analysis for the self-placement data is generally not possible, as the models require these placements to estimate the corresponding latent respondent positions. This section therefore focuses on estimating the ELPD for stimulus placements.

The rstan and loo packages contain functions for estimating ELPDs by performing approximate leave-one-out (LOO) cross-validation using Pareto smoothed importance sampling (PSIS). However, PSIS-LOO only works when all Pareto k values are sufficiently low, or when the number of high values is so low that moment matching can be used on the problematic cases. In practice, this is not always the case, and it rarely seems to be the case with the BAM model.

The hbamr package therefore includes model codes and functions to perform K-fold cross-validation. The function hbam_cv() is similar to hbam() and takes the same arguments, but will perform a K-fold cross-validation for the chosen model. In contrast to hbam(), the default for hbam_cv() is to not allow respondents to have any missing values (allow_miss = 0). The reason is that the cross-validation essentially creates missing values by excluding some data from each run.

A key choice when performing cross-validation is to set K (the number of folds to use), and the default in this package is 10. The function is parallelized so that users can specify a higher number of cores than chains and run chains for different folds simultaneously. To preserve memory, the function extracts the summaries of the log-likelihoods for the held-out data and drops the stanfit objects once this is done. The memory requirements of the function is therefore similar to running a single analysis with one chain per core. The function that splits the data into K folds uses a default seed to produce the same folds each time, unless a different seed is specified.

To perform 10-fold cross-validation for a selection of models, we could run:

elpd_hbam <- hbam_cv(self, stimuli, model = "HBAM")
elpd_hbam_nf <- hbam_cv(self, stimuli, model = "HBAM_NF")
elpd_bam <- hbam_cv(self, stimuli, model = "BAM")

Rather than returning stanfit objects, the function returns a data frame containing the estimated ELPD and its standard error. To compare results across models, we could run:

elpds <- rbind(elpd_bam, elpd_hbam, elpd_hbam_nf)
elpds[order(elpds$ELPD), ]

##            ELPD   SE
## BAM     -5835.7 58.6
## HBAM_NF -5766.9 51.2
## HBAM    -5524.6 48.6

We see that the unpooled BAM model is worse at predicting out-of-sample data than the other models, suggesting it overfits the data. The HBAM_NF model – which does not allow for scale flipping – also performs worse, suggesting it is too restrictive and underfits. The HBAM model outperforms the other two models in this case.

We could also perform cross-validation for a model that does not account for heteroskedastic errors:

elpd_hbam_mini <- hbam_cv(self, stimuli, model = "HBAM_MINI")
elpd_hbam_mini

##              ELPD   SE
## HBAM_MINI -5889.4 47.1

The estimated ELPD is notably higher for this model compared to its heteroskedastic counterpart. This suggest there is a considerable degree of heteroskedasticity in the data – which is not surprising.

Modeling the heteroskedasticity increases prediction accuracy, but it should be noted that this does not necessarily translate into much more accurate estimates of key model outputs. As shown in section on plotting, the results for the HBAM and HBAM_MINI models are very similar. In fact, their estimated respondent positions correlate at .97 and .99, depending on whether we use the posterior means or medians. Their estimated stimulus positions also correlate at .99, but this masks the fact that the HBAM_MINI model places the less well known candidate John B. Anderson further to the left than the HBAM model does. This illustrates that there are some subtle differences that users should be aware of, even if these models tend to produce very similar results.

Limits to Exact Replication

The functions in this package accept a seed argument for all operations that involve a random number generator. Supplying a seed is sufficient to get the exact same results in repeated runs on the same system, but it does not ensure exact replication across systems or software versions. The Stan Reference Manual explains why:

Floating point operations on modern computers are notoriously difficult to replicate because the fundamental arithmetic operations, right down to the IEEE 754 encoding level, are not fully specified. The primary problem is that the precision of operations varies across different hardware platforms and software implementations.

Stan is designed to allow full reproducibility. However, this is only possible up to the external constraints imposed by floating point arithmetic.

In short, running the functions in this package on different systems will likely yield slightly different results. It should be noted, however, that the differences across systems will be minimal and substantively negligible as long as the user obtains a sufficient number of posterior draws.