Their survey compared means across different questions, between universities, across academic ranks, and among academics of different age and gender.
I threw it into Stata.
The questions seemed to load onto a single factor - one factor had an Eigenvalue around 5 with the rest below 1.
So I stuck it into a principal component model to get scores for each respondent for that factor.
The nice bit with a PCF is that the predicted values on the factor are mean zero, standard-deviation 1 by construction. So the coefficients on all the different dummy variables can quickly be understood. A 0.5 coefficient on something means that something is associated with a half a standard deviation increase in the dependent variable.
Higher scores here mean higher academic freedom.
Here are the results from a straightforward OLS with the principal component from the factor analysis over on the left hand side, and the kitchen sink on the right hand side. T-stats below coefficients; stars mean their usual thing.
Principal Component | |
Institution | |
Victoria University of Wellington | |
(Reference category) | |
Otago University | -0.335 |
(3.42)** | |
AUT | -0.516 |
(4.12)** | |
Massey University | -0.380 |
(3.81)** | |
University of Canterbury | -0.141 |
(1.33) | |
Auckland University | -0.262 |
(2.51)* | |
University of Waikato | -0.336 |
(2.74)** | |
Gender | |
Male | |
(Reference Category) | |
Female | -0.103 |
(1.61) | |
Gender Diverse | 0.542 |
(1.90) | |
Field of Study | |
Arts | |
(Reference category) | |
Business | -0.256 |
(2.33)* | |
Creative Arts | -0.272 |
(1.36) | |
Education | 0.384 |
(2.63)** | |
Engineering | -0.487 |
(3.51)** | |
Health Sciences | 0.104 |
(0.97) | |
Laws | 0.002 |
(0.01) | |
Science | -0.123 |
(1.23) | |
Other Area | -0.060 |
(0.48) | |
Rank | |
Lecturer | 0.192 |
(1.78) | |
Senior Lecturer | |
(Reference Category) | |
Ass. Prof | -0.099 |
(1.15) | |
Prof | 0.160 |
(1.93) | |
Senior Management | 0.145 |
(1.01) | |
Other | -0.091 |
(0.82) | |
Age | |
Under 30 | 0.314 |
(1.77) | |
31-45 | |
(Reference Category) | |
46-64 | -0.315 |
(4.27)** | |
65+ | -0.171 |
(1.89) | |
Constant | 0.504 |
(4.27)** | |
R2 | 0.10 |
N | 1178 |
Just a bit of fun - a quick and dirty run through the thing.
Vic Uni comes out well.
Rank doesn't do much controlling for age. Gender doesn't do much.
Folks in Education seem to feel relatively unconstrained; whether that's due to academic freedom or to constraints that match their ideological preferences and so are non-binding is harder to tell.
The Free Speech Union could do well in academia, given what other views there on academic freedom look like.
No comments:
Post a Comment