This markdown includes data from Chemistry students in classes CHM 121, CHM 221, and CHM 222 in Fall 2019 and Fall 2020. Data from the Spring 2020 semester were excluded due to the changes in university classes necessitated by the Covid-19 pandemic. Students who dropped and who completed the class are both included in the data.

1 Class section


Class section participant count:

Fall 2019 Fall 2020
CHM 121 181 262
CHM 221 0 94
CHM 222 0 64

Takeaway:
Students in class section CHM 121 had a significantly higher likelihood of dropping the class when compared to class sections CHM 221 and CHM 222.

2 Gender

2.1 Grades by gender

Students’ average grades are calculated by averaging the grades (0-100) of their 4 or 5 exams.


Takeaway:
No significant difference was found between Male and Female participants’ average grades in their class.


2.2 Goal Grade by gender

At the time of each exam, participants were asked for their goal grade in the class.

goal grade screenshot



Takeaway:
Male students were found to have a significantly higher average goal grade than female students by an average of 1.4 points.


2.3 Prediction Error by gender

Prediction Error is the difference between the grade a student predicted they got on an exam and the actual grade they got on that exam (Actual exam grade - Predicted exam grade). A negative prediction error means that the predicted grade was higher than their actual grade, meaning they overestimated. Participants were asked for their grade prediction on the day of the exam.

goal grade screenshot


Takeaway:
Males had a significantly lower average prediction error by 3 points, meaning there was a greater gap between their predicted grades and their actual grades.

2.4 Drops by gender

Takeaway:
Males and females were found to be equally likely to drop their class.

3 Race

In the plots and analyses below, data from students who identified as “Native American” or “Native Hawaiian or other Pacific Islander” were omitted due to small group sizes (2 students and 1 student, respectively). Data from students who identified as “Other” were not included as a unified group due to the heterogeneity of the races specified.

3.1 Grades by race


Takeaway:
Asian-American students’ average grades were significantly higher than those of other races by an average of 5.4 points.

3.2 Goal Grades by race



Takeaway:
Asian-American students’ average goal grades were significantly higher than those of other races by an average of 3 points.


3.3 Prediction Error by race


Takeaway:
No significant difference in prediction error was found between races.

3.4 Drops by race

Takeaway:
Participants of different races were found to be equally likely to drop their class.

4 Ethnicity

4.1 Grades by ethnicity


Takeaway:
No significant difference was found in the average grades between hispanic/latino and non-hispanic/latino students.


4.2 Goal Grade by ethnicity



Takeaway:
No significant difference was found in the average goal grades between hispanic/latino and non-hispanic/latino students.


4.3 Prediction Error by ethnicity


Takeaway:
No significant difference in prediction error was found between Hispanic/Latino and Non-hispanic/latino participants.

4.4 Drops by ethnicity

Takeaway:
Hispanic/Latino and Non-hispanic/latino participants were found to be equally likely to drop their class.


5 Demographics Summary


Participant count by gender, ethnicity, and race:

Female, Hispanic/Latino Female, Non-hispanic Male, Hispanic/Latino Male, Non-hispanic
African American or Black 6 46 0 7
Asian or Asian-American 2 57 0 19
Caucasian or White 106 184 43 68
Multiracial 5 30 5 9
Native American 2 0 0 0
Native Hawaiian or other Pacific Islander 0 0 1 0
Other 3 3 2 1



6 Analyses Summary


Gender

  1. Grades:
    No significant difference was found between Male and Female participants’ average grades in the class.

  2. Goal Grades:
    Male students were found to have a significantly higher average goal grade than female students by 1.4 points.

  3. Prediction Error:
    Males had a significantly lower average prediction error than Females by an average of 3 points. This means there was a greater gap between Male’s predicted grades and their actual grades, with their actual grades being lower than their predicted grades.

  4. Drop out:
    Males and females were equally likely to drop their class.

Race

  1. Grades:
    Asian-American students’ average grades were significantly higher than those of other races by an average of 5.4 points.

  2. Goal Grades:
    Asian-American students’ average goal grade were significantly higher than those of other races by an average of 3 points.

  3. Prediction Error:
    No significant difference in average prediction error was found between races.

  4. Drop out:
    Participants of different races were found to be equally likely to drop their class.

Ethnicity

  1. Grades:
    No significant difference was found in the average grades between hispanic/latino and non-hispanic/latino students.

  2. Goal Grades:
    No significant difference was found in average goal grades between hispanic/latino and non-hispanic/latino students.

  3. Prediction Error:
    No significant difference in prediction error was found between Hispanic/Latino and Non-hispanic/latino participants.

  4. Drop out:
    Hispanic/Latino and Non-hispanic/latino participants were found to be equally likely to drop their class.

Class section

  1. Grades:
    Students in CHM 221 had a significantly lower average exam grade by an average of 7.5 points when compared to those in CHM 121 (both Fall 2019 and Fall 2020) and CHM 222.

  2. Goal Grades:
    Students’ average goal grades were significantly lower in CHM 221 and CHM 222 when compared to CHM 121 (both Fall 2019 and Fall 2020).

  3. Prediction Error:
    Students’ average Prediction Error were significantly lower in CHM 121 (Fall 2019) when compared to CHM 121 (Fall 2020), CHM 221 (Fall 2020), and CHM 222 (Fall 2020),

  4. Drop out:
    Students in classes CHM 221 and CHM 222 were less likely to drop the class when compared with CHM 121 (both Fall 2019 and Fall 2020).



What factors predict class drop out?

  1. Difference between goal grade and actual grade:
    The difference between a student’s goal grade for the class and their actual average grade was a strong predictor of whether that student dropped the class. Although their average grade alone also significantly predicted drop out, the difference between their grade and their goal was an even stronger predictor.

  2. Class section:
    Class section also predicts drop out. Those in classes CHM 221 and CHM 222 are less likely to drop the class when compared with CHM 121 (both Fall 2019 and Fall 2020).

Do these effects differ based on demographics?

No, ‘goal grade minus actual grade’ and class section predict drop out no matter the gender, race, and ethnicity.