Defining the Problem (1). There are approximately 50 million people in the United who repo...

Defining the Problem (1). There are approximately 50 million people in the United who report having a physical handicap. Furthermore, it is estimated that the unemployment of noninstitutionalized handicapped people between the ages of 18 and 64 is nearly double the unemployment rate of people with no impairment. Thus, it appears that people with disabilities have a more difficult time obtaining employment. One of the problems confronting people having a handicap may be a bias by employers during the employment interview.

The paper “interviewers’ decisions related to applicant handicap type and rater empathy” [Human performance (1990) 3: 57-171] describes a study that examines these issues. The purpose of the study was to investigate whether different types of physical handicaps produce different empathy by raters and to examine whether interviewers’ evaluations are affected by the type of handicap of the person being interviewed.

Collecting Data (2). The researchers videotaped five simulated employment interviews. To minimize bias across videotapes, the same male actors (job applicant and interviewer) and the same interview script, consisting of nine questions, were used in all five videotapes. The script was directed toward average qualifications of the applicant because this type of applicant is the most likely to be susceptible to interview biases. The videotapes differed with respect to type of applicant disability, but all were depicted as permanent disabilities. The five conditions were labeled wheelchair, Canadian crutches, hard of hearing, leg amputee, and nonhandicapped (control).

Each participant in the study was asked to rate the applicant’s qualifications for a computer sales position based on the questions asked during the videotaped interview. Prior to viewing 1ile videotape, each participant completed the Hogan Empathy Scale. The researchers decided to hav each participant view only one of the five videotapes. Based on the variability in scores of rater in previous studies, the researchers decided they would require 14 raters for each videotape in order to obtain a precise estimate of the mean rating for each of the five handicap conditions. Seventy undergraduate students were selected to participate in the study. For each of the five videotape,14 students were randomly assigned to view the videotapes. After viewing each videotape, the participant rated the applicant on two scales: an 11 -item scale assessing the rater's liking of the applicant and a 10 -item scale that assessed the rater’s evaluation of the applicant's job qualifications. For each scale, the average of the individual items form an overall assessment of the applicant. The researchers used these two variables to determine whether different types of physical handicaps are reacted to differently by raters and to determine the effect of rater empathy on evaluations of handicapped applicants.

These are some of the questions that he researchers were interested in:

1. Is there a difference in the average empathy scores of the 70 raters?

2. Do the raters’ average qualification scores differ across the five handicap conditions?

3. Which pairs of handicap conditions produced different average qualification scores?

4. Is the average rating for the control group (no handicap) greater than the average ratings for the groups for all types of handicapped applicants?

5. Is the average qualification rating for the hard-of-hearing applicant group different from the average ratings for the groups whose applicants had a mobility handicap?

6. Is the average qualification rating for the crutches applicant group different from the average ratings of groups whose applicants were either amputees or in wheelchairs?

7. Is the average rating for the amputee applicant group different from the average rating for the wheelchair applicant group?

Summarizing Data (3). The researchers conducted the experiments and obtained the following data from the 70 raters of the applicants. The data in the following table are a summary of the empathy values.

The data in the next table are the applicant qualification scores of the 70 raters for the five handicap conditions along with their summary statistics. (These data were simulated using the summary statistics of the ratings given in the paper. These data are in the HANDICAP file on the CD that came with your book.)

a. Plot the qualification scores in boxplots and discuss what you sec based on these plots and the descriptive statistics shown here.

Analyzing Data, Interpreting the Analyses, and Communicating Results (4). The objective of the study was to investigate whether an inter viewer’s evaluation of a job applicant is affected by the physical handicap of the person being interviewed. Prior to testing hypotheses and making comparisons among the five treatments, we need to verify that the conditions under which the tests and multiple comparison procedures are valid have been satisfied in this study.

---

b. Based on your findings in part (a), does it appear the AOV assumptions hold?

---

c. A probability plot of the residuals is shown here. Does it confirm what you found inpart (b)? Explain.

---

d. Check the equal variance assumption using Levene’s test.

We check the condition of the independence of the data by discussing with the researchers the manner in which the study was conducted. It is important to make sure that the conditions in the room in which the interview tape was viewed remained constant throughout the study, so as not to introduce any distractions that could affect the raters’ evaluations. Also, the initial check that the empathy scores were evenly distributed over the five groups of raters assures us that a difference in empathy levels did not exist in the five groups of raters prior to their evaluation of the applicants’ qualifications. The research hypothesis is that the mean qualification ratings, μis, differ over the five handicap conditions:

H0: μ1 = μ2 = μ3 = μ4 = μ5

Ha: At least one of the means differs from the rest.

The computer output for the AOV table is given here. The following notation is used in the output: Control (C), Hard of Hearing (H), Amputee (A), Crutches (R), and Wheel chair (W). Use the output to draw conclusions.