International Journal of Open Educational Resources
Withdrawal rates
During data analysis , it was noticed that there were a greater number of W grades for OER courses ( n = 57 , 10 % of total grades ) than for non-OER courses ( n = 41 , 7.5 % of total grades ). This was an unexpected observation if it is inferred that lack of access to course materials may trigger withdrawing from the course . A z-test of proportions revealed that OER courses had a statistically significant higher percentage of W grades than non-OER courses ( z = 1.864 , p = . 03 ).
This data were also analyzed by ethnicity . Of 658 white students in the study , 7 % ( n = 46 ) received a W grade . Of 97 black / African American students , 11 % ( n = 11 ) received W grades , as did 11 % of the 253 Hispanic students in the study ( n = 28 ). The difference in proportion of black / African American students receiving W grades compared to white students did not exceed a 95 % confidence level ( z = 1.615 , p = . 053 ). However , the difference in proportion of Hispanic students receiving W grades compared to white students was significant ( z = 2.358 , p = . 009 ) ( refer to Table 3 ).
Table 3 . Withdrawal rates by ethnicity across all courses
A-F W Total % Withdrawals Asian 73 7 80 9 %
Black / African American
86 11 97 11 %
Hispanic 225 28 253 11 % White 612 46 658 7 % Total 996 92 1088 8 %
Discussion
This study aimed to answer two research questions :
1 . Is there a significant difference in student achievement for students in OER sections compared to sections of the same course taught by the same instructor in a previous semester using a commercial textbook ?
2 . Is there a significant difference in student achievement for Pell recipients , non-white students , first-time students or part-time students in OER sections compared to previous semester sections taught by the same instructor using a commercial textbook ?
Consistent with the majority of OER research on academic performance , the results of this study found
54