Wednesday, March 7, 2012

In Just 1 Hour: Reducing the Achievement Gap

Today we bring you a post by another amazing guest blogger, Michelle Rheinschmidt. Michelle is a graduate student at Berkeley and her guest post highlights some of the astounding effects that 1 hour and a few posters can have on academic and career outcomes! 
We can probably think of a time when concerns about “fitting in” affected our behavior in adolescence, but what about in adulthood?  

New environments, such as starting college or a new job, make people worry about whether they will be accepted by others. These concerns can be amplified when people belong to underrepresented, stereotyped, or devalued social groups (e.g., women in STEM fields, ethnic minority students). Research suggests that underrepresented students worry about whether they belong in college settings, and these concerns interfere with achievement. In fact, experimental research aimed at reducing belonging concerns has been shown to reduce race-based achievement gaps by over 50%.

What did this research look like?
Researchers Gregory Walton and Geoffrey Cohen randomly assigned White and African American college freshman to receive a social-belonging intervention (i.e., treatment group) or to serve in a “non-treated” control group.  The social-belonging intervention was predicted to have the largest effect for African American students, for whom belonging concerns are especially relevant given numerical underrepresentation and academic stereotypes. Students assigned to the treatment group read testimonials from older students about how their concerns about “fitting in” at college diminished with time. These testimonials described belonging concerns as temporary and common to all students (i.e., not just ethnic minority students).  Students not receiving the treatment (i.e., control group) read testimonials about developing “social-political” views over time.
This 1-hour intervention changed the course of African American students’ college trajectories. African American students in the treatment condition showed a greater sense of belonging in college and better GPAs than those in the control condition. The GPAs of African American students in the treatment condition rose over the 3-year observation period, ending up .14 grade points away from those of White students, as opposed to .29 in the control condition – a 52% reduction. White students were relatively unaffected by the treatment.To help explain these effects, the authors looked at how students made sense of the day-to-day challenges they faced in college, especially during the transition. They found that African American students who received the treatment were less likely to view daily adversity as a sign that they did not belong in college. That is, students generated explanations for academic and social challenges that were not tied to having a “place” in college. In post-study surveys, students were not aware that they had co-opted the intervention’s message or of its reoccurring effects. What is truly remarkable about this social-belonging intervention is its profound impact on minority student achievement, given its brief and inexpensive nature.

Computer Science Stereotypes
Just as college trajectories can be affected by belonging concerns, so too can important career choices. Social psychologist Sapna Cheryan and her colleagues explored the reasons why women may opt out of careers in computer science. They suspected that a perceived lack of “fit” with the people and things in one’s environment may preclude interest in particular fields. To test this hypothesis, they manipulated the room in which male and female students who had not yet chosen a major completed a survey. The survey included questions about their interest in computer science. The room had items pre-rated as either stereotypical and “nerdy” (e.g., Star Trek poster, electronics) or non-stereotypical (e.g., nature poster). Women were more interested in the field of computer science when they completed the survey in the non-stereotypical versus stereotypical room. Men’s interest in the field did not differ by room. The authors concluded that women’s “ambient belonging”, or sense of fit with the people and things around them, was threatened in the stereotypical room. This study suggests that even subtle environmental cues (e.g., a poster) can make people question whether they belong in a certain domain or field of study.

Feelings of belonging and acceptance have implications not just during our awkward teen years, but also in adulthood, in ways as profound as shaping college trajectories and influencing career decisions. These outcomes are especially important for fostering diversity and even social mobility. Research studies, such as those described above, hold promise for making school and work environments more inclusive. 

Can you believe a 1-hour intervention can have that dramatic of an effect? Have you had situations where you didn’t feel like you belonged? What did you do to change that?

The Articles: 

  • Walton, G., & Cohen, G. (2007). A question of belonging: Race, social fit, and achievement. Journal of Personality and Social Psychology, 92 (1), 82-96 DOI: 10.1037/0022-3514.92.1.82
  • Walton GM, & Cohen GL (2011). A brief social-belonging intervention improves academic and health outcomes of minority students. Science (New York, N.Y.), 331 (6023), 1447-51 PMID: 21415354
  • Cheryan, S., Plaut, V., Davies, P., & Steele, C. (2009). Ambient belonging: How stereotypical cues impact gender participation in computer science. Journal of Personality and Social Psychology, 97 (6), 1045-1060 DOI: 10.1037/a0016239

 Michelle Rheinschmidt is a Ph.D. candidate in the social and personality psychology program at UC Berkeley. Her research explores how membership to different social groups (ethnic and social class groups, in particular) influences interpersonal behavior and educational experiences.


  1. Very interesting. Can you tell us more about the 3 year observation period and the characteristics of the students in the study? It is surprising (and promising) the effect of the intervention in the minority student's performance. Although I wonder about their environment for the 3 year-period, and how the researchers did to control for external factors that could be causing an increase sense of belonging in the students in the treatment condition.

    1. Michelle RheinschmidtMarch 12, 2012 at 3:46 PM

      Thank you for your comment, MF. I appreciate your interest!

      The sample was drawn from two cohorts of incoming students at a selective university. Students were randomly assigned to the treatment and control groups to minimize the influence of other student characteristics (which should be randomly distributed across conditions). There were two groups against which the treated students were compared: 1) students in the control group and 2) students from the same class year and ethnicity who did not participate in the study (but for whom anonymous GPA data was obtained). The use of random assignment and two comparison groups helps isolate the effect of the intervention.

      The intervention seemed to affect students in a few different ways (described below), which then yielded cumulative effects over time. In particular, African American students who received the treatment construed daily events differently than those who had not. In the week following the intervention, control participants' feelings of belonging were contingent on daily adversity, whereas treated students' feelings of belonging were more stable, even in the face of adversity. At the 3-year follow up, African American students in the treatment group reported more confidence in their belonging, less stereotype accessibility, and less self-doubt, compared to African Americans in the control group. These differences in event construal and information processing help explain how the treatment affected students and their college outcomes.

      I hope this helps clarify some of the psychological processes behind the group differences. I highly recommend the articles above. Thank you for reading!