Psychological researchers can study important and contemporary issues, including an issue that came to the forefront in 2017--sexual misconduct by people in power. This article in the Washington Post summarizes several examples of research on sexual harassment. The journalist wrote about studies that identified which people are more likely to harass, and also which situations are more likely to cause people to sexually harass.
The journalist first described a self-report scale designed to identify men who are more likely to sexually harass women. Read about the scale's format:
As one of the pioneers in the study of sexual harassment, Pryor developed a test in 1987 to measure a man’s tendency to harass. Called the “Likelihood to Sexually Harass” scale, Pryor’s test has become a cornerstone of research on sexual harassers.
His test consists of 10 scenarios. Imagine that you are an executive hiring a new secretary, one scenario starts out. A female candidate explains she desperately needs the job and looks at you in a way that possibly conveys she is attracted to you. How likely are you to give her the job? Offer the job in exchange for sexual favors? Ask her to go to dinner to discuss the job?
(There's a video mockup of the scale associated with the journalist's article)
In Chapter 5, you learned about how a self-report measure should, ideally, be investigated for its reliability and validity. A self-report scale of tendency to sexually harass should be shown to be reliable and valid.
a) Which forms of measurement reliability are probalby most relevant for this scale: Internal, inter-rater, or test-retest?
Next, read what the journalist writes about the sexual harassment scale:
Over the years, Pryor — a psychologist at Illinois State University — and others have used socially engineered situations in laboratories to study how well the test predicts people's behavior.
b) Reflect on what the self-report scale is supposed to measure. Now, consider what kind of "socially engineered situations in laboratories" would show that this self-report scale has criterion validity? That is, what kinds of behaviors should high scores on the self-report scale be correlated with?
Next, consider this finding from Pryor's research with the self-report scale of propensity to sexually harass:
And over time, they’ve identified these factors as the most distinctive in harassers: a lack of empathy, a belief in traditional gender sex roles and a tendency toward dominance/authoritarianism.
c) Do the relationships (described above) count as convergent validity, discriminant validity, or neither?
In addition to describing this self-report scale, the journalist also summarizes a second, separate line of research on sexual harassment; this one investigating the impact of holding a powerful position on perceptions of sexual interest.
A particularly eye-opening 2011 study found that people in leadership often pick up phantom sexual signals from subordinates that aren’t really there.
The experiment designed by Jonathan Kunstman and Jon Maner took 78 adults and paired them with a member of the opposite sex. Those pairs were assigned a Lego-building project, with one person put in charge of the other. In private interviews at the end of the project, those who were appointed leaders were much more likely to have perceived sexual interest from their subordinates, even when the subordinate said in surveys that they had no sexual interest at all.
When researchers studied video of most pairs interacting, they found the leaders much more likely to act on that misperception, touching the subordinate's leg or engaging in eye gazing.
d) Consider the design of this study. Assuming that participants within each pair were randomly assigned to be leaders or subordinates in the Lego task, what kind of experiment was this? (posttest only? pretest/posttest? repeated measures? concurrent measures?)
e) Many experiments have more than one dependent variable (DV). What were the DVs in this study?
f) Can the results support the claim that "Being in a leadership position during a Lego exercise caused people to misperceive sexual interest and to touch their subordinates' legs"? Apply the three causal criteria of covariance, temporal precedence, and internal validity.
g) Comment on the external validity of the "leader/subordinate" Lego study. To what extent does this study extend to situations in the "real world", such as sexual harassment allegations against powerful people such as Harvey Weinstein or Matt Lauer? Do the researchers intend to have their study inform such situations?