Why WAGES? A Demonstration.
Women have made tremendous strides in scientific fields since the 1970s. Women work hard and know what it takes to succeed. Most men do not want to be unfair and, in fact, know that gender equity benefits them, too, when their female partners and family members can advance and be paid fairly. Despite women’s hard work and men’s good will, however, gender-related inequities in opportunity, advancement, and salary persist in nearly every industry and profession studied.
Three Scenarios
Think about how you would handle each of the following situations. Then click on Food for Thought for additional ideas.
Scenario A
You are chair of Department Y’s search committee for one of three new tenure-line positions. You are looking over the CVs of the top 10 candidates to recommend three to be brought in for interview. Three of the ten candidates are women. You want to be sure that all candidates are fairly evaluated.
How can you be sure that both search and selection process are fair?
Food for Thought
- We know that when candidates are compared on a case-by-case basis it is easy to overlook or explain away qualifications that don’t match our expectations. University of Michigan’s ADVANCE program has developed a “candidate evaluation tool” that helps search committees maintain the objectivity they strive for: http://sitemaker.umich.edu/advance/stride
- Ensuring equity in a search starts long before the long short-list is developed. Getting a large, qualified, and diverse pool of applicants means more than advertising in the usual places and contacting colleagues for nominees. Instead, an effective search needs to tap into networks that don’t overlap greatly with your own.
- What proportion of the candidate pool nationally is comprised of women? Does three out of 10 represent at least the availability of potential women applicants? If not, it may be useful to go through applications again to determine whether a high quality candidate from an underrepresented group may have been overlooked.
- Watch out for “diffusion of responsibility.” It may be tempting to think that, because there are 3 searches underway, a more diverse pool of qualified applicants will automatically occur. Not so. Each individual search committee must work to cast a wide net.
Scenario B
You have been tapped to be on the selection committee for the Prestigious Publication Prize of an important professional society. Candidates for the prize can be nominated by others or be self-nominations. The other members of the committee are respected colleagues you’ve known for years and you feel good about how well the group works together. You are looking over the final list of award candidates.
How can you be sure that the selection process is fair?
Food for Thought
- Ensuring equity in a search starts long before the review of nominations gets underway. Ensuring the best nominations means more than advertising in the usual places and contacting colleagues for nominees. Instead, an effective search needs to tap into networks that don’t overlap greatly with your own.
- Understand that members of some groups are more willing to self-nominate than are others and this can influence both who is in the pool of nominees and what the committee sees as fitting the norm for nominees.
- Case-by-case comparisons of individual nominees are difficult to do objectively. The power of stereotypes and other cognitive shortcuts, combined with the apples vs. oranges nature of piecemeal information, is a sure recipe to lead us to ignore information that doesn’t fit with pre-existing knowledge.
Scenario C
You learn that in a department very much like your own, one of the senior men is paid somewhat more than a senior woman. Both were hired by University of Great Importance in the same year when both were just completing two year post-docs. The man has a few more publications than the woman, but in all other respects she has the same or better qualifications.
Is it possible to rule out salary discrimination by taking a good, close look at the two faculty members’ CVs?
Food for Thought
- We know that when candidates are compared on a case-by-case basis it is easy to overlook or explain away qualifications that don’t match our expectations. Scenario C points out that the man has a few more publications than the woman and so it is easy to assume that quantity of publications is the driving force behind the salary difference. There are two problems with this.
- First, of course, is that quantity, by itself, tells us nothing about the quality, significance, impact of the publications or the individual’s role in the project if the publication is co-authored.
- Second, and perhaps less obvious, Scenario C also notes that the woman “in all other respects…has the same or better qualifications.” Which qualifications are better?
- In order to determine whether inequity exists, we need to go beyond the individual case and begin to look at aggregated data to see whether there are, in fact, patterns that suggest inequity. In the case of individuals, the comparisons need to be made within the same domain, not across domains. So, in Scenario C, we won’t learn much from comparing his publication list with her grants.
Why good intentions aren't enough.
Good intentions and efforts to be objective are not enough.
If you are like most of us, you are confident of your ability to be objective and impartial when it comes to weighing and comparing the qualifications of colleagues in your area of expertise and your discipline more generally.
However, a keen eye and good intentions alone are not enough to neutralize powerful—and quite natural—information processing biases shared by all humans. In fact, for Scenarios A and B, actions to ensure gender fairness must start long before the final set of job candidates/prize nominees is on the table.
What’s going on here? The answer, it turns out, is in the way that all humans process information. Much of the time we rely on cognitive shortcuts. On the plus side, these shortcuts make it possible for us to juggle quantities of complex information without becoming bogged down in detail. On the negative side, however, cognitive shortcuts can lead us to overlook information or come to premature conclusions—even when we aim to be objective.
How people process social information.
How we process social information, especially information about other people, can be heavily influenced by stereotypes, a form of cognitive shortcut that entails quick and unconscious generalization about an individual based on his or her group membership.
Other kinds of cognitive short-cuts can trip up our objectivity, too. For example, when asked to think of examples of a given category like “vegetables” or “movies,” or “prominent scientists,” we are more likely to think of examples we’ve recently encountered or that are more commonly encountered.
Stereotypes, both positive and negative, can have powerful effects, even when we do not explicitly believe them. Research in scientific psychology repeatedly shows that everyone—regardless of the social groups we belong to—is influenced by stereotypes. We are influenced even by stereotypes of our own group.
Cognitive short-cuts impair objectivity.
These unconsciously deployed stereotypes underlie the persistence of patterns of gender inequity in academic engineering and science. Prejudice and blatant discrimination do still occur in work environments. Today, however, most gender discrimination happens “under the radar.” See a sample of WAGES items for an illustration.
Even when we know about the power of stereotypes and cognitive short-cuts to impair our objectivity, it takes work to organize evaluative information in such a way to reduce our susceptibility to those biasing influences. Crosby and colleagues have shown that even just the arrangement of the data can disguise or reveal a pattern of inequity. Click here for a demonstration.
To learn more about the power of cognitive short-cuts and our difficulty in neutralizing their biasing effects, see: http://www.understandingprejudice.org/iat/, http://www.understandingprejudice.org/asi/, and http://mitworld.mit.edu/video/80/.
Can you spot the inequity?
Take a couple of minutes to examine Table 1 below and determine whether there is systematic salary inequity that favors one team or the other. Level refers to job, with higher numbers indicating the more advanced, demanding position. Years refers to years with Company Z.
Table 1. Base information about levels, years, and salaries in Company Z
| |
Average Green Team managers |
Average White Team managers |
| Department |
Level |
Years |
Salary |
Level |
Years |
Salary |
| A |
9 |
6 |
$41,000 |
12 |
10 |
$67,000 |
| B |
10 |
9 |
$49,000 |
12 |
13 |
$71,000 |
| C |
11 |
8 |
$52,000 |
13 |
17 |
$83,000 |
| D |
12 |
8 |
$56,000 |
13 |
16 |
$80,000 |
| E |
12 |
16 |
$68,000 |
14 |
19 |
$89,000 |
| F |
12 |
20 |
$74,000 |
8 |
5 |
$44,000 |
| G |
13 |
15 |
$70,000 |
9 |
8 |
$52,000 |
| H |
13 |
19 |
$76,000 |
11 |
5 |
$55,000 |
| I |
14 |
17 |
$78,000 |
12 |
7 |
$63,000 |
| J |
14 |
20 |
$81,000 |
12 |
14 |
$73,000 |
Your decision?
- No visible inequity exists.
- There is salary inequity favoring Green Team managers
- There is salary inequity favoring White Team managers
Now examine Table 2 and determine whether there is systematic salary inequity that favors one team or the other.
Table 2. Base information about levels, years, and salaries in Company Z
| |
Average Green Team managers |
Average White Team managers |
| Department |
Level |
Years |
Salary |
Level |
Years |
Salary |
| A |
9 |
6 |
$41,000 |
8 |
5 |
$44,000 |
| B |
10 |
9 |
$49,000 |
9 |
8 |
$52,000 |
| C |
11 |
8 |
$52,000 |
11 |
5 |
$55,000 |
| D |
12 |
8 |
$56,000 |
12 |
7 |
$63,000 |
| E |
12 |
16 |
$68,000 |
12 |
14 |
$73,000 |
| F |
12 |
20 |
$74,000 |
12 |
10 |
$67,000 |
| G |
13 |
15 |
$70,000 |
12 |
13 |
$71,000 |
| H |
13 |
19 |
$76,000 |
13 |
17 |
$83,000 |
| I |
14 |
17 |
$78,000 |
13 |
16 |
$80,000 |
| J |
14 |
20 |
$81,000 |
14 |
19 |
$89,000 |
Your decision?
- No visible inequity exists.
- There is salary inequity favoring Green Team managers
- There is salary inequity favoring White Team managers
If you are like most readers, it was much easier to spot inequity in Table 2, even though the information for both teams is identical in both tables!
Inequity is easy to spot in Table 2 because when we look across departments, 9 of 10 department pairings reveal that disadvantaged managers (Green Team) have higher qualifications, but lower salaries than advantaged managers (White Team).
When the same information is rearranged, however, inequity is harder to spot. Table 1 contains the same information as Table 2, except that the five bottom rows for advantaged managers are switched with the five top rows. In this case, relying on side-by-side comparison masks inequity. Both Table 1 and Table 2 show that Company Y pays less to Green Team managers who have the same qualifications as White Team managers.
If you remain skeptical, just compute the average Level, Years, and Salary columns for the two groups of managers.
The Research Basis
Crosby and colleagues, as well as other researchers, have shown in numerous studies over the years that people are less likely to detect discrimination when information is presented as individual cases (i.e., piecemeal form) or when group data are arranged in a way that masks the relation between qualifications and salary. When the individual cases are aggregated so that idiosyncratic differences in qualifications can be averaged or the relation between qualifications and salary are lined up, people readily identify inequities.
*Adapted from Rutte, C.G., Diekmann, K. A., Polzer, J. T., Crosby, F. J., & Messick, D. M. (1994). Organization of information and the detection of gender discrimination. Psychological Science, 5, 226-231 and Crosby, F., Clayton, S., Alksnis, O, and Hemker, K. (1986). Cognitive biases in the perception of discrimination: The importance of format. Sex Roles, 14, 637-646.
