Tag Archives: Tests

Problems in the pipeline: Stereotype threat and women’s achievement in high-level math courses

1 Star2 Stars3 Stars4 Stars5 Stars (No Ratings Yet)
Loading...

Submitted by Jaclyn Sadicario

Article Reference

Good, C., Aronson, J., & Harder, J. A. (2008). Problems in the pipeline: Stereotype
<p style="padding-left: 30px;">threat and women's achievement in high-level math courses. <em>Journal of Applied </em></p>
<p style="padding-left: 30px;"><em>Developmental Psychology</em>, <em>29</em>, 17-28.
</p>

Article DOI

http://dx.doi.org/10.1016/j.appdev.2007.10.004

Summary of Article

<strong>Background</strong>
<ul>
<li>APA Goals for Psychology “Exhibit respect for members of diverse groups with sensitivity to issues of power, privilege, and discrimination.”</li>
<li>There is a performance gap between men and women on mathematics tests.</li>
</ul>
<ul>
<ul>
<li>Negative stereotypes can undermine women's performance on mathematics tests.</li>
<li>Laboratory studies have confirmed stereotype threat; however, relevance in the  “real world” on women's math test performance have not been tested.</li>
<li>This study tested how much stereotype threat affected math test performance in a group of advanced calculus students.</li>
</ul>
</ul>
<strong>Participants </strong>
<ul>
<li>N= 157 advanced calculus students at a large, public university, 67% (100) men, 33% (57) women.
<ul>
<li>Overall: 57% White, 20% Asian, 11% Hispanic, <1% African American, 7% “other”</li>
</ul>
</li>
</ul>
<strong>Materials & Procedure</strong>
<ul>
<li>They designed & piloted a calculus test, aligned with what the students were learning / GRE math.</li>
<li>Students were told this was a review / practice test & that they’d get extra credit for it based on their score (administered at the end of the semester).</li>
<li>Two conditions
<ul>
<li>(ST) Top sheet with “this test will measure your ability” statement (control- stereotype threat condition)</li>
<li>(NST) Top sheet with “this test will measure your ability” then a statement that there are no gender differences in math ability.</li>
</ul>
</li>
<li>Randomized packets were distributed with:
<ul>
<li>Test, demographics survey, post-test questionnaire, and debriefing form.
<ul>
<li>Post-test questionnaire asked how they felt about the test.</li>
</ul>
</li>
</ul>
</li>
</ul>
<strong>Results</strong>
<ul>
<li>NST women outperformed ST women
<ul>
<li>also outperformed both NST & ST men</li>
</ul>
</li>
<li>NST participants were more accurate than ST participants</li>
<li>NST women significantly more accurate than ST women & ST men</li>
<li>Men were more confident than women about their performance regardless of condition</li>
<li>Men and women do not differ on course grades based on gender <strong>HOWEVER</strong>:
<ul>
<li>Course grades under predicted test scores for NST women.</li>
</ul>
</li>
</ul>
<strong>Discussion</strong>

Even women at the upper ends of the ability distribution in college who opt to enroll in the most difficult math courses can be vulnerable to the effects of negative stereotypes.

Discussion Questions

  1. 1. In the chapter on testing, Forsythe talks at length about guidelines for test construction but does not spend much time talking about the atmosphere in which the test is taken. Based on this article, what do you think he should add to an upcoming chapter?
  2. 2. There was a finding that despite performing better, women in the non-threat condition and the women in the threat condition had equal amount of confidence in their work (lower confidence than "male" students). Have you had experience helping students build confidence in their work?
  3. 3. What kind of intervention do you think would be effective in our student populations at VCU? What might be different about our populations than a small, liberal arts school? Sub-question: Do you think take home tests / online tests would help combat stereotype threat? Why?/Why not?

Online Academic Integrity

1 Star2 Stars3 Stars4 Stars5 Stars (No Ratings Yet)
Loading...

Submitted by Athena Cairo

Article Reference

Mastin, D. F., Peszka, J., & Lilly, D. R. (2009). Online academic integrity. Teaching of Psychology, 36, 174-178. DOI: 10.1080/00986280902739768

Article DOI

10.1080/00986280902739768

Summary of Article

In this study, the authors investigated the extent to which students generally would tend to cheat on an online extra-credit assignment, and whether the time of the semester or signing an honor pledge might moderate these tendencies.

Background

In some older surveys of college students (1, 2), between 40 and 83% of students reported engaging in academic dishonesty at some point in their college career. Between 2%-13% of students reported having cheated in a traditional lecture course in which they were currently enrolled (3). Although slightly more than half of faculty and students think that it would be easier to cheat in an online setting (4), only 3% of students admitted to cheating in online courses (5). Additionally, reports suggest schools that have honor codes tend to have lower rates of cheating among the student body (6). In light of these points, the authors conducted an experiment to investigate the rates of cheating in their student body, and the effect of writing an honor code with the assignment (experimental condition), as well as the time of the semester on cheating rates (moderating variable). Specifically, the researchers hypothesized: 1. Being asked to agree to an honor pledges will discourage cheating 2. Students are more likely to cheat later in the semester than earlier in the semester

Methods

Participants were 439 undergraduate students taking an Intro to Psych class. Participants were tested over the course of three different time points, September 2005 (n = 141), May 2005 (n = 124), and May 2006 (n = 174).

Participants were told they could receive up to 10 bonus points for participating in the study, and were told it was a pilot study of a motor task. Upon signing up to participate, P’s were assigned to one of three pledge conditions: no pledge, check-mark pledge and typed-out pledge.

P’s accessed the motor task online, which was designed to be especially difficult (so students would have an incentive to cheat). P’s were told their points would be tied to their performance on the motor task, but als0 that the page could not track their performance on the task.

P’s completed the motor task or hitting a computer key when the correct number appeared on the screen. At the end of the task, participants reported their number of hits in a text box, allowing them to possibly cheat by over-reporting their successful hits. P’s were debriefed 7 days after participation.

Results

  • Fourteen percent of participants cheated by over-reporting their hits. Overall, participants in the group reported better performance than they obtained; t(438) = –5.37, p < .05, d = .26

  • Honor pledge conditions had no effect on cheating frequencies across all P’s, nor did it predict greater severity of cheating among those who did cheat.
  • Participants were twice as likely to cheat at the end of the semester than at the beginning χ2(2, N = 423) = 6.41, p <.05, Cramer’s V =.12. End vs. beginning of the semester did predict greater severity of cheating.

References:

(1) Bunn, Caudill, & Gropper, 1992; (2); Davis, Grover, Becker, & McGregor, 1992; (3) Kerkvliet & Sigmund, 1999; (4) Kennedy, Nowak, Raghuraman, Thomas, & Davis, 2000; (5) Grijalva, Nowell, & Kerkvliet, 2006; (6) McCabe, Trevino, & Butterfield, 2002

Discussion Questions

  1. Many students and faculty think it’s easier to cheat online than in a traditional lecture course– how could you as an instructor help prevent cheating on different types of assignments (e.g. research paper, tests, online quizzes) in both an online and a traditional lecture course?
  2. Do you think that having students write/sign an honor pledge helps prevent cheating at VCU? Would this maybe depend on different classroom contexts or types of assignment?
  3. Additionally, how can we help students not feel as stressed and compelled to cheat toward the end of the semester?

    Bonus question: Say you find a student who severely cheats or plagiarizes an assignment– what would you do? What if the student was someone you liked or knew was going through difficult circumstances?