Estimate standard errors from conjoint design

This function allows you to estimate a model on a design of experiment that you have not yet used to collect data. This allows you to learn about deficiencies in your design of experiment and also assess the sample size needed to achieve parameter precision levels before you go out and use the design to collect data. The function fills out the survey with random choices and estimates a model. It does this multiple times with an increasing number of observations, set by the nbreaks argument. While the coefficients in those models are meaningless, the standard errors on the coefficients are informative. The example below estimates 10 separate models and then plots the standard errors against the number of observations. In this example, assume that the yogurt data was not a completed survey but rather a blank design of experiment with no observed choices.

sampleSizer(
  survey,
  parNames = NULL,
  parTypes = NULL,
  interactions = FALSE,
  nbreaks = 10,
  randPars = NULL,
  options = list(message = FALSE)
)

Arguments

survey	The choice survey data frame exported from the makeSurvey() function.
parNames	A vector of the names of the parameters to be estimated in the model. Must be the same as the column names in the survey argument.
parTypes	A vector determining the type of each variable: "c" for continuous, or "d" for discrete. Continuous variables will be linearly coded whereas discrete variables will be dummy coded with one level removed for identification. Defaults to NULL, in which case all parameters are coded as "d" for discrete.
interactions	A logical value for adding interactions between all parameters in parNames. Defaults to FALSE.
nbreaks	The number of different sample size groups.
randPars	A named vector whose names are the random parameters and values the distribution: 'n' for normal or 'ln' for log-normal. Defaults to NULL.
options	A list of options to control the model estimation.

Value

Returns a data frame of the standard error values for different sample sizes.

Examples

if (FALSE) {
library(conjointTools)

# A simple conjoint experiment about apples, with one attribute (price)
# modeled as continuous

# Make the design of experiment
doe <- makeDoe(
    levels = c(3, 3, 3),
    varNames = c("price", "type", "freshness"),
    type = "full"
)

# Make the survey
survey <- makeSurvey(
    doe       = doe,  # Design of experiment
    nResp     = 1000, # Total number of respondents (upper bound)
    nAltsPerQ = 3,    # Number of alternatives per question
    nQPerResp = 6     # Number of questions per respondent
)

# Compute sample sizes
results <- sampleSizer(
    survey   = survey,
    parNames = c('price', 'type', 'freshness'),
    parTypes = c('c', 'd', 'd'), # Set continuous vs. discrete variables
    nbreaks  = 10
)

# Preview results
head(results)

# Plot results
library(ggplot2)

ggplot(results) +
    geom_point(aes(x = size, y = se, color = coef),
               fill = "white", pch = 21) +
    scale_y_continuous(limits = c(0, NA)) +
    labs(x = 'Number of observations',
         y = 'Standard Error',
         color = "Variable") +
    theme_bw()
}