|
approach |
data |
sample |
analysis |
strength of evidence |
chain of reasoning |
cost of planning |
cost of execution |
cost of analysis |
1 |
quasi-experimental, multi-institutional |
quant & qual |
8 institutions, 217 subjects |
statistical & qualitative |
medium |
good presentation. Some flaws |
high |
high |
high |
2 |
experimental |
numeric |
72 males |
statistical |
low |
weak |
med-high |
med |
low |
3 |
survey |
scales (open questions) |
?? |
compilation |
LOW |
--- |
low |
low |
low |
4 |
psychometric experiment (2x2 mixed measures) |
numeric (scales) |
90 |
statistical (factor analysis) |
patchy |
patchy |
high |
medium |
medium |
5 |
quasi-experimental pre- & post- test |
numeric |
?? 46-ish ?? |
basic stats |
patchy/low |
weak, gappy & thin |
low |
low |
low |
6 |
census (longitudinal) |
numeric & qual |
48 unis 85 courses (% population? departments?) |
compilation |
strong |
few conclusions drawn |
very low (bugger all) |
very high (shitload} |
low |
7 |
on-line survey |
qual & numeric & grades |
40/200 |
“I put them into a spreadsheet” (basic stats) |
weak (But served its purpose) |
average |
low (3 afternoons) |
very low |
low |
8 |
quasi-experimental survey (validated instruments) pre & post test |
numeric |
~270 (/400) |
statistical |
OK, but limited (based on averages) |
moderate conclusions explicit unexposed assumptions good |
high (part of larger endeavour. This paper a hitchhiker) |
moderate |
med-low |