Mini research update

Without further ado, a few research updates:

January 2017 Psychological Bulletin:

- Why is there better gender balance in some STEM fields than others?  Possible reasons these authors cite include masculine cultures, lack of early experience, and lower self-efficacy.

- There appears to be a strong relationship between emotional intelligence and other personality factors, particularly trait EI.


January 2017 Journal of Applied Psychology:

- Sackett & Lievens argue that a modular perspective on selection procedures --in other words, breaking selection procedures down into their design components such as response format -- allows for insights beyond a holistic view.  Read this quite fascinating article here.


Reminder: the Journal of Personnel Assessment and Decisions is free, available here.

Research update

Several research updates this time, including fascinating studies of how computers can assist with interview training, and how a brief writing exercise can lower stereotype threat for women:


International Journal of Selection and Assessment, December, 2016:

- International support for the cultural intelligence scale

- Looking to improve applicant interview performance? Maybe a computer can help.

- This study found that time lag and g-loading are important factors impacting re-testing results

- Status-seeking seems to be an important individual difference when looking at self-presentation behaviors, including exaggeration and faking in job search

- Development and validation of a 360-degree measure of leadership personality


Personnel Psychology, Winter 2016:

- Do CEOs significantly impact firm performance?  This study found evidence that they do.

- More evidence that the assumption that performance is normally distributed is questionable

- A more accurate correction for range restriction is presented, and an example analysis indicates the relationship between the Big 5 and job satisfaction may be greater than previously believed



Journal of Organizational Behavior, November, 2016 (which includes several articles devoted to the importance of theory in organizational sciences):

- Are self-focused or other-focused recruiting advertisements more effective?  This study suggests it depends not only on cultural differences but individual regulatory focus



Journal of Applied Psychology, October, 2016:

- A suggestion for improving meta-analytic structural equation modeling



Journal of Applied Psychology, November, 2016:

- A fascinating study of how having women compose a brief written description of their personal values can help ameliorate stereotype threat in competitive environments

2014 Research of the Year (+ research update)

Happy New Year!  As I've done in previous years, I present below the research articles I ran across in this area that I think were the most impactful and/or important of 2014.  But first, let's catch up on two issues:

First, the Winter issue of Personnel Psychology:

- Situational judgment tests have been shown to be useful for measuring interpersonal skills, but beware: levels of "angry hostility" moderate that relationship.  (Is there a happy hostility?)

- When hiring leaders, should you look for those that have a busy home life, or be wary of them?  In this fascinating study, the authors found that leader family-to-work conflict negatively impacts followers in that it can increase their burnout. However, family-to-work enrichment increased follower engagement through leader engagement.  So the answer is, as usual, not simple: home/family life can be a good thing for followers if it makes the leader more engaged; but if the home/family life is increasing burnout, the leader may pass that along to others.  So it would seem it all depends on how the individual is handling their life outside of work!

Let's look at the November issue of the Journal of Applied Psychology:

- Are men or women perceived as better leaders?  According to this meta-analysis, it depends on how you ask the question.  If you limit the question to other-ratings, women are rated significantly higher.  But if you look at self-ratings, men rate themselves significantly higher.  Which leads to the next question: is it a biological perception or a gender perception, and if the latter, what traits are the most important?

- An intriguing study of how applicant confidence interacts with and can be altered by the recruitment experience, in this case among recruits for the U.S. military.

- Next, a study of employment and job search efficacy.  Not surprisingly, within-person frequency of job search behavior correlated with job offers; interestingly, the relationship between perceived job search progress and efficacy beliefs were moderated by beliefs of internal attribution.

- Last but not least, more evidence of the importance of defining the criteria when predicting job performance.  In this meta-analysis, the researchers found more support for personality traits out-predicting cognitive ability in predicting counterproductive work behavior, that the two predictors are approximately equal in predicting organizational citizenship behaviors, and that cognitive ability outperforms personality when predicting task and overall performance.  So do you want high task performance, OCBs, or do you want to avoid CWB?  :)  (of course the situation is even more complicated depending on whether you're looking at individual, team, leader performance, over what period of time, etc.)


Okay, on to the awards!  Without further ado, here are my nominations for Research of the Year for 2014:

1) Important advancements in our understanding of weight-based discrimination at work: Vanhove & Gordon.

2) A study of applicants posting faux pas on their social networking sites: Roulin.

3) Two important looks at assessments delivered remotely via mobile devices: Arthur, Doverspike, Munoz, Taylor, & Carr, and Morelli, Mahan, & Illingworth.

4) Two fascinating looks at personality at work: Judge, Simon, Hurst, & Kelley; and Wille & De Fruyt

5) An excellent study of how effective staffing and training practices impact firm-level flexibility and adaptability: Kim & Ployhart.

6) An important study of the movement of impactful I/O researchers to business schools: Aguinis, Bradley, & Brodersen.

7) The relationship between conscientiousness and job performance is more accurately described as curvilinear: Carter, Dalal, Boyce, O'Connell, Kung & Delgado

Finally, honorable mention to two great developments in 2014:  the change of some publishers to making access to articles more affordable, and the announcement of an additional journal, the Journal of Personnel Assessment and Decisions.


I'm continually amazed at the quality of thought and research in our area and the passion and practicality you exhibit.  Here's to an amazing 2014 and more in 2015!

January 2012 research update

Okay, we've got a lot of ground to cover so let's jump right in...

- For you stats fans: Estimating ordinal reliability for Likert-type and ordinal response data

- And more: Effect size: Don't just report it, describe it

- And still more: A confidence interval approach to testing comparability of groups

- Shifting gears: CWBs: There's more than one type

- Femininity and military leadership

- Increasing HR's strategic participation

- HR practices do have an influence on organizational attractiveness

- IQ scores may be influenced by test motivation (in low-stakes settings)

- Learning agility is an important factor in identifying high-potentials

- Slicing and dicing the Big Five

- Extraversion, intelligence, and genetics

- Apparently being agreeable won't make you rich (especially if you're a man)

- Need scientists or engineers? Consider innovativeness and internal LOC

- What if you're looking for leaders of scientists or engineers? Consider three important criteria.

- Increase diversity cues to increase recruitment website stickiness

- Does transformational leadership relate to proactive behavior? Depends on the situation.

- Core self evaluations: More complicated than you might think

- No, seriously, it's much more complicated and involves general versus specific measures

- Speaking of complicated, the relationship between personality traits (Big 5) and performance isn't as simple as many make it out to be

That's all for now! See you in February.

Should you hire more women for your teams?

Should you hire more women for the teams within your organization? You might think so after reading an article in the June 2011 Harvard Business Review. It's an interview with the authors of some research that came out last year in Science. In fact this hiring strategy has even been suggested based on this research.

But let's take a deeper look.


The takeaways from the HBR article (and the published study) suggest:

- there is a "collective intelligence" factor (c) that is related to team success

- this (c) factor out-predicts team success compared to the average team intelligence score, the highest intelligence score among the team members, or other logical factors such as group cohesion and satisfaction

- the (c) factor is primarily related to the average social sensitivity of the team members, the equality of distribution of turn-taking during team conversation, and...(drumroll please) the proportion of females in the group

In the studies, the authors had nearly 700 individuals (one assumes students? the subjects aren't described) participate in teams of two to five on a variety of tasks, such as completing puzzles, brainstorming, and negotiating. At the end of the session they had them complete the criterion task--in the first study a video game of checkers against a computer opponent, in the second, an architectural design task.

So what did they find? As Kai Ryssdal would say, let's do the numbers:

- There did seem to be some general factor that predicted a significant amount of variance in the criteria (43% and 44% respectively).

- (c) seems to be related, at least to a small amount, with both average individual intelligence (r=.15) and maximum member intelligence (r=.19), but the authors stress higher correlations with average social sensitivity of group members (r=.26), variance in the number of speaking turns (r=-.41), and proportion of females in a group (r=.23), although the latter was largely a result of the women scoring much higher on the measure of social sensitivity, which the authors stress came out on top in terms of unique prediction power.

- the instrument used to measure social sensitivity, the "Reading the mind in the eyes" test, has subjects identify the emotion being displayed by a set of eyes (reminiscent of some emotional intelligence tests I've seen). It would be interesting to see how well other measures of social sensitivity (e.g., body language, tone of voice) predicted team decision making, and tie this with other research that has shown emotional intelligence measures predicting team performance.

- The standardized regression coefficients (betas) for (c) were .51 and .36 for the two criteria, substantially above average member intelligence (.08, .05) and maximum member intelligence (.01, .12).

- (c)'s relationship with performance on the various tasks in Study 1 varied pretty widely, from .38 to .86. This, combined with the differential prediction of the criteria, suggests (c) as conceptualized may be more useful for predicting performance on certain group tasks. It's worth noting that the lowest correlation was with brainstorming--a task that requires less team interaction.

- The authors do not say what instrument was used to measure individual intelligence. This may or may not matter.


There are some important lessons here:

1. As is often the case, the farther we get from the actual publication, the more important it is to view the interpretation with caution. In this case, I believe some writers have over-emphasized and over-played the "flashy" result (more women on team -> better decisions) and failed to consider things like effect sizes or relationships among variables. What I'm more interested in is why the women scored higher.

2. From this research the concept of a collective intelligence factor does seem promising (and has been the subject of other recent popular publications). In reality this line of research is old as well as thriving, and includes such well-researched concepts as groupthink as well as several lines of research around what makes an effective team.

3. It is important to remember that job performance is multi-faceted. We know this from (among other things) previous research that has shown intelligence tests do a better job predicting task performance than contextual performance, where non-cognitive tests are at their best (this fact has interesting implications for the study that is the subject of this post). The results of this study remind us to carefully consider what behaviors we're hiring for.

4. It's studies like this that, when improperly analyzed, muddy the waters of our profession. Using this research to say you should hire more women is like saying you should hire more Whites than Blacks because they tend to score higher on intelligence tests. Aside from the obvious discriminatory intent, this is just plain bad decision making: it over-emphasizes differences at the group level and assumes that you have clear evidence that intelligence tests are highly correlated with performance in the job you are hiring for (and similarly valid tests with smaller mean group differences are unavailable).

I have to give the authors credit for going beyond gender as a causal factor in predicting team performance and looking for root relationships, and for not leaping to conclusions like organizations should hire more women, but instead focusing our attention on the implications for team development (they suggest electronic collaboration tools may increase collective intelligence).

This type of press is great for getting us to talk more about what matters. Let's just make sure when we do so we start with the research and consider all the important angles.

July 2010 J.A.P.

A new round of journals is out, so let's start with the June issue of the Journal of Applied Psychology.

First up, Schleicher et al. looked at whether there were demographic differences in how much candidate scores improved upon retesting. Turns out there were several. Whites showed larger improvements than Blacks or Hispanics on several assessments, particularly on written tests. Women and applicants under 40 showed greater improvements than men and applicants 40+. Implications? In some situations allowing applicants to retest may exacerbate adverse impact.

Next, an important piece by Aguinis et al. (that you can read here) about test bias. This follows on the heels of the June IOP articles on the same topic and seems to represent a resurgence of interest in a topic that seemed dormant. In this article the authors report the results of a very large Monte Carlo simulation (billions and billions of data points) where they found that if bias is measured using slope-based techniques, it's likely to go undetected, and intercept-based bias favoring minority group members is likely to be found when in fact it does not exist. This study, combined with points made in the IOP article suggest that some of the "established" conclusions regarding test bias may not be as solid as we thought.

Third, for those of you interested in differential functioning (of items or scales), you should check out the piece by Adam Meade where he presents a taxonomy of potential differential functioning effect sizes and also describes a software program created for computing the indices and graphing differential functioning.

Next, a piece by Wang et al. on locus of control. Importantly, they found that when locus of control (LOC) is specific to work-related issues, there are stronger correlations between LOC and work-related criteria such as job satisfaction and commitment. Similarly, when LOC is defined more broadly to include non-work issues, there are some stronger correlations with non-work criteria such as life satisfaction. Implications? Much like research on personality items, specifying a work-related context would seem to increase the predictive power of LOC measures.

Last but not least, an important article on counterproductive work behavior (CWB) and organizational citizenship behavior (OCB) by Spector, et al. CWB and OCB seem like they should be opposites of each other--one demonstrated by disengaged, unhappy workers, the other by engaged, happy ones--right? Not so fast. The authors report the results of an experiment that suggest that the concepts are unrelated and do not necessarily have opposite relationships with other variables. The authors also recommend that when measuring these behaviors, frequency of performance be used rather than level of agreement.

June 2010 IJSA

The summer journal season continues with the June 2010 issue of the International Journal of Selection and Assessment. Take a deep breath, there's a lot of stuff packed into this issue:

- Roth et al. provide evidence that women outperformed men on work sample exams that involved social skills, writing skills, or a broad array of KSAs. To the extent that an employer is trying to avoid discriminating against female applicants, this provides support for work sample usage.

- In a study of managers in Taiwan, Tsai et al. show that the most effective way an applicant can make up for a slip in an interview is to apologize (vs. attempting to justify or use an excuse).

- Jackson et al. strive to add some clarity on task-based assessment centers

- Blickle & Schnitzler provide evidence of the construct and criterion-related validity of the political skill inventory

- Colarelli et al. studied how racial prototypicality and affirmative action policies impact hiring decisions. Results of a resume review indicated more jobs were awarded to black candidates as racial prototypicality and affirmative action policy strength increased, but stronger AA policies decreased the percentage of minority hires attributed to higher qualifications.

- In my personal favorite article of the issue, Karl et al. found in a study of U.S. and German students that those low on conscientiousness (especially), agreeableness, and emotional stability were more likely to post "Facebook Faux Pas". This provides some support for employers who screen out applicants based on inappropriate social networking posts. I'll talk more about this in my upcoming webinar.

- Denis, et al. provide support for the NEO PI-R's ability to predict job performance in two French-Canadian samples.

- Bilgiç and Acarlar report results of a study of Turkish students and perceptions of various selection instruments. Interviews were rated most highly and there were some differences in terms of privacy perceptions depending on the goal orientation of the student.

- Trying to figure out how to hire better direct support professionals (e.g., those providing long-term residential care or care to those with disabilities)? Robson, et al. describe the development of a composite predictor composed of various measures (e.g., agreeableness, numerical ability) that predicted performance, satisfaction, and turnover.

- Ahmetoglu et al. provide support for using the Fundamental Interpersonal Relationship Orientations-Behaviour (FIRO-B) to predict leadership capability.

- Ispas et al. describe results of a study that showed support for a nonverbal cognitive ability measure (the GAMA) in predicting job performance in two samples.

- Last but not least, in another win for context-specific assessments, Pace & Brannick show how a measure of openness to experience tailored to specific work outpredicted the comparable general NEO PI-R scale. IMHO this is how personality measures will eventually become more prominent and accepted as pre-hire assessments.

March 2010 J.A.P.

The March 2010 issue of the Journal of Applied Psychology is out. Let's take a look:

Do women make better leaders? According to a study by Rosette and Tost, it varies with how success is attributed, the level of the position, perceptions of double-standards, and expectations. So the answer? A very solid "it depends."

Who should determine SJT scoring? Motowidlo and Beier suggest in their research study that situational judgment test (SJT) scoring keys based on input from subject matter experts (SMEs) contribute differentially to the prediction of job performance compared to keys based on general knowledge about trait effectiveness. What does this mean? That your ability to predict performance using SJTs depends in part on who is determining the scoring, and getting SME input may boost the effectiveness.

Do Americans work to live or live to work? Based on an analysis from Highhouse, et al., it's looking more and more like the former.

Need more evidence of the value of confirmatory testing? Naquin, et al. performed three experimental studies that demonstrated higher levels of lying when using email compared to pen and paper.

Do you like your leaders proactive? According to research conducted in China by Ning et al., you're not alone.

Finally, a slight correction to an article by Ilies et al. published last July on the relationship between personality and organizational citizenship behaviors.

Latest IJSA: Emotional intelligence, multiple-choice formats, and lots more

The March 2010 issue of the International Journal of Selection and Assessment (IJSA) is out, and the research covers a wide variety of recruitment and assessment topics as well as being truly international:

Unproctored internet-based testing (UIT) response distortion may be less than we fear (sample included cognitive and personality measures)

What factors are most important to organizations when choosing a test? This study suggests applicant reaction, cost, and diffusion of the test type in the field.

Personality (esp. core self evaluation) is related to the type of work preferred, and hence P-O fit

Career site features may differentially attract men and women

Corporate images do matter when it comes to organizational attractiveness

Who uses job-search websites and how to improve them (the sites, not the people)

Support for performance-based (as opposed to self-report) measures of emotional intelligence

Work samples, interviews, and ability tests perceived best by employees (why? because they work, say the participants)

...and last but definitely not least:

A "2 of 5" multiple-choice format seems superior than traditional "1 of 6" (you just have to make sure you can score them that way!)

Want better prediction? Gather more data.

That's the bottom line from a study in the November 2009 issue of the Journal of Applied Psychology.

Oh & Berry looked at how adding personality ratings from peers and supervisors added incremental validity to self-ratings using a five-factor model measure. What were the results? Increases of 50-74% in operational validity across personality facets. They also looked at differential prediction of task and contextual performance (unfortunately those results weren't reported in the abstract). Bottom line? If you're using a personality assessment for promotions, strongly consider gathering data from co-workers.

Speaking of self-presentation, in the same issue Barrick et al. report the results of a meta-analysis of how self-presentation tactics (e.g., appearance, non-verbal behavior) impact interview ratings and later job performance. Results? "What you see in the interview may not be what you get on the job and...the unstructured interview is particularly impacted by these self-presentation tactics." An important reminder of how who the candidate seems to be impacts your assessment, and another reason to collect multiple sources of data.

There are a number of other great articles in this issue, such as:

How Major League Baseball CEO personalities impact important outcomes (like, um, winning).

How SJT and biodata measures add to the prediction of college student performance.

How personality scale validities change over time among a group of medical students.

Differences among letters of recommendation in academia between genders.

Is recruiting using SNS discriminatory?

I keep reading/hearing about how recruiting using social networking sites (SNS) opens employers up to discrimination lawsuits because of who uses the sites. For the most part, this just plain isn't true.

A recent Pew study is the latest to show that when it comes to using SNS like Facebook, MySpace, and LinkedIn, you really should have one primary demographic concern when it comes to ensuring a diverse candidate pool: age.

Not gender, at least not in traditional sense. While four years ago SNS users tilted slightly male (55%), the balance has essentially flipped today (54% female).

Not race, there simply do not appear to be generalizable differences in racial groups when it comes to these sites (in fact I've seen some data that suggest the user base on these sites is more diverse)--but things change, and this may vary with particular sites, so keep an eye on this one.

But when it comes to age, SNS users are disproportionately younger than the overall Internet population. In the words of the Pew report, "[this] doesn't mean that more older adults aren't flocking to SNS--they are--but younger adults are ALSO flocking to the sites, so the overall representation of the age cohorts in the SNS user population has actually gotten younger."

One demographic difference I don't see a whole lot about: disability status. Are individuals with disabilities more/less likely to use SNS? I think that's an important question we need to address if we're truly trying to diversity our candidate pools.

The red clothing effect

A little fun for our Friday...

I'm not a big fan of interviews. Particularly ones that are unstructured (e.g., different questions for different candidates, no rating scales, etc.).

Why? Aside from the fact that research has shown them to be much less predictive of job success, they're problematic because most people think they are above average interviewers (they're obviously not), that they're particularly good at picking up things like deception and lies (once again...they're not), and that interviews are done easily and quickly (they shouldn't be).

Another reason interviews are tricky is they're susceptible to all kinds of perceptual errors. Some of the more common ones include:

- The "Halo" effect: something about the candidate biases the way you see other things about them. My favorite example is having positive feelings toward someone because they went to your alma mater.

- The contrast effect: your opinion of a candidate is biased because they followed a particularly good or bad candidate.

- The fatigue effect: the way you evaluate candidates changes over the course of a day or week because you get tired of interviewing.

This is just a sample of cognitive biases that enter into the interview process. Other typically non-job related factors come also into play, such as someone's height.

Now we may have to add the red clothing effect. A recent study of undergraduates at the University of Rochester found that the color red, compared to other colors, led heterosexual men to find women more attractive (it had no effect on female participants' perception of other females).

The researchers validated this (albeit with small samples) using a variety of experiments, including digitally altering the shirt color of the same image. They even looked at other factors such as willingness to ask out on a date (hint for straight women: chose red over blue).

The silver lining for us is that color had no impact on perceptions of likability, kindness, or intelligence. Still, it's something to be aware of that could potentially have important consequences. After all, remember what happened to Neo.

Too fat or too thin? You may not get hired.

Job candidates that are either too fat or too thin may have a more a difficult time getting hired than those in the middle weight ranges according to a study by Swami, et al. reported in the most recent issue of the Journal of Applied Social Psychology.

Weighting in line
The authors found that when men were asked to rate a variety of female pictures for either a management position or for providing help (N=30 and 28, respectively), they were less likely to hire or help women with body mass indices (BMI) over 30 or under 15. Those with a slender body (BMI = 19-20) were most likely to be hired or helped. This shouldn't be surprising, given that studies have consistently linked physical attributes, including weight, with employment decisions, but it's certainly a reminder to watch your biases when evaluating candidates!

Predict-ability
In another article, Truxillo et al. found a relationship between cognitive ability and the ability to accurately judge one's performance on an employment test. Using a video-based situational judgment test of customer service skills, the authors found that those with high cognitive ability were able to predict their performance while those with low cognitive ability were not. Practical implications? Providing thorough test feedback may be particularly important for candidates lower in cognitive ability as they may be more likely to be surprised (and dismayed) by the results. This means providing information prior to the test as well as afterward (e.g., how it was developed, how it is scored, how you can improve your performance).

Working IT
In a third study, Johnson, et al. found gender and ethnic group differences in how IT careers are perceived as well as in self-efficacy related to IT. Using data from 159 African- and 98 Anglo-Americans, the authors found that African American men reported higher levels of IT self-efficacy than all other groups, whereas Anglo women reported the lowest levels. In addition, Anglos had more negative stereotypes of IT professionals than did African Americans. This study had a small sample size, but the implication is that how people see their own ability related to an occupation, as well as how they perceive those in it, influences their career choices. This will in turn impact your applicant demographics as well as your recruiting success.

The rest
There are some other interesting reads in here, including:

When emotional displays of leaders may increase follower performance

How to give performance feedback

Self-perceptions of ethical behavior

2007 SIOP Conference: Highlights, Part 2

This is the third in a series of posts about the 2007 SIOP Conference. In Part 1 I talked about some of the new products out there and in Part 2 I went over some of the research that was presented. In this post I'll point out some more research that you may find interesting...

Legal risks and defensibility factors for employee selection procedures

Posthuma, Roehling, and Campion analyzed nearly 600 federal district court cases and came up with some very interesting results:

- Employers are most likely to win (by far) when defending tests of math or mechanical ability. Employers also fare well when defending assessments of employment history and interviews.

- Employers did worst when defending physical ability tests and medical examinations. Tests of verbal ability and job knowledge were also more likely to result in a plaintiff win.

Predicting Internet job search behavior and turnover

Using a sample of 110 nurses in Texas, Posthuma et al. found using longitudinal survey data that (among other things) Internet job search behavior was related to turnover--folks weren't just surfing for fun. This suggests that organizations need to pay close attention to job searching behavior among employees; not necessarily to curtail it but instead to figure out why high performers want to leave.

Gender differences in career choice influences

After analyzing survey data from nearly 1,400 fourth-year medical students from two U.S. schools, Behrend et al. found a gender difference in preferred career: specifically, female medical students valued "opportunities to provide comprehensive care" when choosing a specialty much more than men. This is consistent with other work that has showed women to be more "relationship-oriented" than men when it comes to choosing a career.

Portraying an organization's culture through properties of a recruitment website

In this study of 278 undergraduate students, Kroustalis and Meade found that inclusion of pictures on a website that were intended to portray a certain organizational culture did so--but only for certain cultural characteristics. Specifically, pictures that implied a culture of either innovation or diversity had the intended effect--but pictures representing a team orientation did not. Interestingly, "employee testimonials" designed to emphasize these cultural aspects failed to do so for any of the three aspects studied. Finally, individuals who perceived a greater fit between themselves and the organization (in terms of the three cultural aspects) reported being more attracted to the organization.

Recruiting solutions for adverse impact: Race differences in organizational attraction

Last but definitely not least, Lyon and Newman gathered data from nearly 600 university students on their reactions to 40 hypothetical job postings...and came away with some very interesting results. For example:

- Conscientious individuals were more likely to apply to postings that explicitly stated a preference for conscientious applicants.

- Conscientious individuals were more likely to apply to postings that described the company as results-oriented.

- Black applicants with higher cognitive ability were more likely to respond to ads seeking conscientious individuals while White applicants with higher cognitive ability were less likely to do so.

- When a company was described as innovative, Black applicants high on conscientiousness were more likely to apply; this was not the case for White applicants.