scholarly journals The validity of student evaluation of teaching: Is there gender bias?

2019 ◽  
Vol 21 (3) ◽  
Author(s):  
Milica Maričić ◽  
Aleksandar Đoković ◽  
Veljko Jeremić
Keyword(s):  
Gender Bias ◽  
Regression Models ◽  
Assessment Tool ◽  
Student Evaluation ◽  
Great Success ◽  
Linear Regression Models ◽  
Student Evaluation Of Teaching ◽  
Independent Variables ◽  
Level Of Satisfaction ◽  
Evaluation Of Teaching

Student evaluation of teaching (SET) has steadily, but surely, become an important assessment tool in higher education. Although SET provides feedback on students level of satisfaction with the course and the lecturer, the validity of its results has been questioned. After extensive studies, the factor which is believed to distort the SET results is gender of the lecturer. In this paper, Potthoff analysis is employed to additionally explore whether there is gender bias in SET. Namely, this analysis has been used with great success to compare linear regression models between groups. Herein, we aimed to model the overall lecturer impression with independent variables related to teaching, communication skills, and grading and compare the models between genders. The obtained results reveal that gender bias exists in certain cases in the observed SET. We believe that our research might provide additional insights on the interesting topic of gender bias in SET.

Prédiction de la composition corporelle post-mortem de cerfs de Virginie (Odocoileus virginianus) à partir de mesures morphologiques et massiques

1997 ◽  
Vol 75 (11) ◽  
pp. 1790-1795 ◽  
Author(s):  
Chantal Bois ◽  
Michel Crête ◽  
Jean Huot ◽  
Jean-Pierre Quellet
Keyword(s):  
Odocoileus Virginianus ◽  
Regression Models ◽  
Carcass Composition ◽  
Chemical Analyses ◽  
Linear Regression Models ◽  
Mass Measurements ◽  
Composition Corporelle ◽  
Independent Variables ◽  
Multiple Linear Regression Models ◽  
Field Stations

Morphologic and mass measurements were taken on 24 complete white-tailed deer (Odocoileus virginianus) carcasses of varying ages and both sexes in southern Quebec. Each carcass was divided into three parts (skin, viscera, rest) to determine water, protein, fat, and ash content by chemical analyses. Fat content of carcasses varied between 0.8 and 17.4%. Multiple linear regression models were selected to predict carcass composition from morphologic and mass measurements. Two situations were considered: measurements taken at the laboratory on whole animals and measurements taken at field stations on eviscerated carcasses provided by hunters. All selected models can be applied to any deer without taking into account age or sex; they include 1 – 4 independent variables. For whole animals, adjusted R2 of models varied between 0.99 (water) and 0.89 (ash); models developed for field stations were less precise, the lowest R2 values being 0.82 and 0.73 for ash and fat, respectively. These models can be useful for research and management purposes.

Sign in / Sign up

Export Citation Format

Share Document