“Gender Differences in Academic Performance”

“Gender Differences in Academic Performance�

Before you go into Descriptive analysis:

1. Select a random stratified sample from the target population (students enrolled in 3rd

semester undergraduate degree). (stratum: major)with n= N/2 Your dataset includes

GPA – grade point average per last semester- calculated as Grade (Numerical Equivalent) x Credits = Grade Points/Number of credits taken last semester

Cumulative GPA- cumulative grade point average – calculated as Total Grade Points / Total Credits taken since 1st semester

Age- measured in years

Gender-coded: 1=male, 2=female

Major- coded: 1=Media & Communications , 2= History , 3= International Relations PART 1: Participants

2. Start writing your report describing the demographics about your respondents. Represent the proportion of males and females enrolled in Media & Communication, History and International Relations major. Any difference? (hint: gender by Major(frequency table). Comment on how many people (count) from your sample were either Media & Communication or History students? State what major was represented the most and the least. Also add the information that you know about the AGE (for full sample): relevant measures of central tendency and dispersion given the shape of distribution. Include stacked bar plot for gender by major and histogram and boxplot for Age for each gender separately.

3. create 4 equal Cumulative GPA groups ordered them from the lowest to the highest and label the four new groups ascendingly as poor, average, above average, excellent. Find how many History students performed better than value corresponding to the 60th percentile inclusive for their respective Cumulative GPA distribution.

Find the value corresponding to 10th percentile (exclusive and inclusive) in Female International Relation’s cumulative GPA data and interpret it correctly. Now look at the gender frequencies across Cumulative GPA groups . Is there any difference in frequencies of males and females across those 4 groups? Generate a frequency table and state what proportion of the males out of full sample was classified in above average and poor GPA groups. Out of all media students, how many students were included in average and above average categories (+ state cumulative frequency).

Bonus: Display an ogive polygon for Cumulative GPA for males and then females (2 separate polygons = x- axis= class boundaries , y-axis cumulative frequencies from 0- 100%). Compare the two (only state if there is any difference in frequency of GPA across classes) and conclude what GPA class interval seems to be occupied the most in male and female data. (what interval is represented the most= the sharpest increase in ogive curve)

PART 2: Descriptive statistics

4, write a full sample descriptive statistics of GPA variable (include graphical

representation, relevant measures of central tendency and dispersion). Also display a boxplot and comment if there are any outliers or not.

5, write a split sample descriptive statistics of GPA scores based on gender! Compare mean, median and mode between Males and Females’ GPA data. Look at the SD, range, variance, CoV in both groups and articulate in which the scores varied more. Evaluate the normality and shape of distribution (symmetry and flatness plus Shapiro’s Wilk) of both groups separately. Are there any outliers detected in any of the groups? If yes, comment (include Case numbers and values in male and female data) if not, calculate the upper and lower limit for detection of outliers using IQR method for both males and females. Please include Whisker boxplot and histogram for both genders.

6, create a composite variable called Academic Index (Raw Cumulative GPA and Age combined). For full sample, write the maximum and minimum z-score and interpret it appropriately. State what is the mean and SD.

BONUS tasks: Evaluate the relative proportion (in %) of those scoring more and less than 1 standard deviation from the average Performance Index score. Find what was the proportion of females and then males scoring more than 1 SD above the mean. Compare the difference in proportion of males and females scoring more than 1SD above the Academic Index average score.