[banner]

Summary and Analysis of Extension Program Evaluation in R

Salvatore S. Mangiafico

Goodness-of-Fit Tests for Nominal Variables

Goodness-of-fit tests are used to compare proportions of levels of a nominal variable to theoretical or expected proportions.  Common goodness-of-fit tests are G-test, chi-square, and binomial or multinomial exact tests.

 

In general, there are no assumptions about the distribution of data for these tests.  However, the results of chi-square tests and G-tests can be inaccurate if cell counts are low.  A rule of thumb is that all cell counts should be 5 or greater for chi-square- and G-tests.  For a more complete discussion, see McDonald in the “Optional Readings” section for details on what constitutes low cell counts.

 

One approach is to use exact tests, which are not bothered by low cell counts.  However, if there are not low cell counts, using G-test or chi-square test is fine.  G-test is probably technically a better test than chi-square.  The advantage of chi-square tests is that your audience may be more familiar with them.

 

G-tests are also called likelihood ratio tests, and “Likelihood Ratio Chi-Square” by SAS.

 

Appropriate data

•  A nominal variable with two or more levels

•  Theoretical, typical, expected, or neutral values for the proportions for this variable are needed for comparison

•  G-test and chi-square test may not be appropriate if there are cells with low counts in them

 

Hypotheses

•  Null hypothesis:  The proportions for the levels for the nominal variable are not different from the expected proportions.

•  Alternative hypothesis (two-sided): The proportions for the levels for the nominal variable are different from the expected proportions.

 

Interpretation

Significant results can be reported as “The proportions for the levels for the nominal variable were statistically different from the expected proportions.”

 

Packages used in this chapter

 

The packages used in this chapter include:

•  EMT

•  DescTools

•  ggplot2

 

The following commands will install these packages if they are not already installed:


if(!require(EMT)){install.packages("EMT")}
if(!require(DescTools)){install.packages("DescTools")}
if(!require(ggplot2)){install.packages("ggplot2")}

 

Goodness-of-fit tests for nominal variables example

 

As part of a demographic survey of students in this environmental issues webinar series, Alucard recorded the race and ethnicity of his students.  He wants to compare the data for his class to the demographic data for Cumberland County, New Jersey as a whole


Race                Alucard’s_class   County_proportion
White               20                0.775
Black                9                0.132
American Indian      9                0.012
Asian                1                0.054
Pacific Islander     1                0.002
Two or more races    1                0.025
-----------------  ---               ------
Total               41                1.000


Ethnicity           Alucard’s_class   County_proportion
Hispanic             7                0.174
Not Hispanic        34                0.826
-----------------   ---               -----
Total               41                1.000


Exact tests for goodness-of-fit

 

Race data


observed = c(20, 9, 9, 1, 1, 1)
expected = c(0.775, 0.132, 0.012, 0.054, 0.002, 0.025)

library(EMT)

multinomial.test(observed, expected)

  ### This can take a long time!

Exact Multinomial Test, distance measure: p

    Events    pObs    p.value
   1370754       0          0


### A faster, approximate test by Monte Carlo simulation

observed = c(20, 9, 9, 1, 1, 1)
expected = c(0.775, 0.132, 0.012, 0.054, 0.002, 0.025)

library(EMT)

multinomial.test(observed, expected,
                 MonteCarlo = TRUE)


Exact Multinomial Test, distance measure: p

    Events    pObs    p.value
   1370754       0          0


Ethnicity data


x =  7
n = 41
expected = 0.174

binom.test(x, n, expected)


Exact binomial test

number of successes = 7, number of trials = 41, p-value = 1


G-test for goodness-of-fit

 

Race data


observed = c(20, 9, 9, 1, 1, 1)
expected = c(0.775, 0.132, 0.012, 0.054, 0.002, 0.025)

library(DescTools)

GTest(x=observed,
      p=expected,
      correct="none")

   ### Correct: "none" "williams" "yates"


Log likelihood ratio (G-test) goodness of fit test

data:  observed
G = 46.317, X-squared df = 5, p-value = 7.827e-09


Ethnicity data


observed = c(7, 34)
expected = c(0.174, 0.826)

library(DescTools)

GTest(x=observed,
      p=expected,
      correct="none")

   ### Correct: "none" "williams" "yates"


Log likelihood ratio (G-test) goodness of fit test

G = 0.0030624, X-squared df = 1, p-value = 0.9559


Chi-square test for goodness-of-fit

 

Race data


observed = c(20, 9, 9, 1, 1, 1)
expected = c(0.775, 0.132, 0.012, 0.054, 0.002, 0.025)

chisq.test(x = observed,
           p = expected)


Chi-squared test for given probabilities

X-squared = 164.81, df = 5, p-value < 2.2e-16


Ethnicity data


observed = c(7, 34)
expected = c(0.174, 0.826)

chisq.test(x = observed,
           p = expected)


Chi-squared test for given probabilities

X-squared = 0.0030472, df = 1, p-value = 0.956


Multinomial test example with plot and confidence intervals

 

This is an example of a multinomial test that includes a bar plot showing confidence intervals.  The data is a simple vector of counts, like the above example. 

 

For a similar example using two-way count data which is organized into a data frame, see the "Examples of basic plots for nominal data" section in the Basic Plots chapter.

 

Walking to the store, Jerry Coyne observed the colors of nail polish on women’s toes (Coyne, 2016).  Presumably because that’s the kind of thing retired professors are apt to do.  He concluded that red was a more popular color but didn’t do any statistical analysis to support his conclusion.


Color of polish    Count
Red                19
None or clear       3
White               1
Green               1
Purple              2
Blue                2


We will use a multinomial goodness of fit test to determine if there is an overall difference in the proportion of colors (multinomial.test function in the EMT package).  The confidence intervals for each proportion can be found with the MultinomCI function in the DescTools package.  The data then needs to be manipulated some so that we can plot the data as counts and not proportions.

 

Note here that the expected counts are simply 1 divided by the number of treatments.  In this case the null hypothesis is that the observed proportions are all the same.  In the examples above, the null hypothesis was that the observed proportions were not different than expected proportions.  It’s two ways to think of the same null hypothesis.

 

The confidence intervals for each proportion can be used as a post-hoc test, to determine which proportions differ from each other, or to determine which proportions differ from 0.


nail.color = c("Red", "None", "White", "Green", "Purple", "Blue")
observed   = c( 19,    3,      1,       1,       2,        2    )
expected   = c( 1/6,   1/6,    1/6,     1/6,     1/6,      1/6  )


library(EMT)

multinomial.test(observed,
                 expected)

   ### This may take a while.  Use Monte Carlo for large numbers.


Exact Multinomial Test, distance measure: p

    Events    pObs    p.value
    237336       0          0


library(DescTools)

MCI = MultinomCI(observed,
                 conf.level=0.95,
                 method="sisonglaz")

MCI


            est    lwr.ci    upr.ci
[1,] 0.67857143 0.5357143 0.8423162
[2,] 0.10714286 0.0000000 0.2708876
[3,] 0.03571429 0.0000000 0.1994590
[4,] 0.03571429 0.0000000 0.1994590
[5,] 0.07142857 0.0000000 0.2351733
[6,] 0.07142857 0.0000000 0.2351733


Nail.color = factor(nail.color,
                    levels=unique(nail.color))

      ### Order the levels, otherwise R will alphabetize them


### For plot, Create variables of counts, and then wrap them into a data frame

Total = sum(observed)
Count = observed
Lower = MCI[,'lwr.ci'] * Total
Upper = MCI[,'upr.ci'] * Total

Data = data.frame(Count, Lower, Upper)

Data


  Count Lower     Upper
1    19    15 23.584853
2     3     0  7.584853
3     1     0  5.584853
4     1     0  5.584853
5     2     0  6.584853
6     2     0  6.584853


library(ggplot2)

ggplot(Data,                 ### The data frame to use.
       aes(x     = Nail.color,
           y     = Count)) +

    geom_bar(stat = "identity",
             color = "black",
             fill  = "gray50",
             width =  0.7) +

    geom_errorbar(aes(ymin  = Lower,
                      ymax  = Upper),
                      width = 0.2,
                      size  = 0.7,
                      position = pd,
                      color = "black"
                      ) +
    theme_bw() +
    theme(axis.title = element_text(face = "bold")) +

    ylab("Count of observations") +
    xlab("Nail color")


image

Bar plot of the count of the color of women’s toenail polish observed by Jerry Coyne while walking to the store.  Error bars indicate 95% confidence intervals (Sison and Glaz method).


Optional readings

“Small numbers in chi-square and G–tests” in McDonald, J.H. 2014. Handbook of Biological Statistics. www.biostathandbook.com/small.html.

 

References

Coyne, J. A. 2016. Why is red nail polish so popular? Why Evolution is True. whyevolutionistrue.wordpress.com/2016/07/02/why-is-red-nail-polish-so-popular/.

 

“Chi-square Test of Goodness-of-Fit” in Mangiafico, S.S. 2015a. An R Companion for the Handbook of Biological Statistics, version 1.09. rcompanion.org/rcompanion/b_03.html.

 

“Exact Test of Goodness-of-Fit” in Mangiafico, S.S. 2015a. An R Companion for the Handbook of Biological Statistics, version 1.09. rcompanion.org/rcompanion/b_01.html.

 

“G–test of Goodness-of-Fit” in Mangiafico, S.S. 2015a. An R Companion for the Handbook of Biological Statistics, version 1.09. rcompanion.org/rcompanion/b_04.html.

 

“Repeated G–tests of Goodness-of-Fit” in Mangiafico, S.S. 2015a. An R Companion for the Handbook of Biological Statistics, version 1.09. rcompanion.org/rcompanion/b_09.html.