Class Notes (806,631)
Canada (492,365)
SOAN 2120 (388)

October 30.docx

4 Pages
Unlock Document

University of Guelph
Sociology and Anthropology
SOAN 2120
David Walters

October 30, 2012 Next Thursday is the midterm…. Brad (201 Blackwood) M at 2:30-4:30 Leanna – F at 10-11 Stephanie F at 3:30-4:30 Rachel- W at 10-11 NEXT CLASS WE WILL BE WORKING WITH SPSS…… GET IT. The Solomon Four-groups Design Combines elements of both the post-test only and the pretest-posttest control group designs. R O1 (BEFORE TREATMENT) X(TREATMENT) O2 (OBSERVED AGAIN) R 03 X O4 R X O5 R O6 Non-Experimental Research Designs - random assignment or even matched pairs assignment is not practical - non-equivalent control group design (field experiments) - new teaching technique (unlikely that classes split up for purposes of experimentation) - minimum wage program affects the likelihood that welfare recipients might find jobs o ‘test program’ in one community versus other communities as ‘control groups’ – problem is the lack of random assignment - the people have to be randomly assigned to those groups – come from different neighbourhoods and went to different schools, similar to each other but different to each other when compared across schools Research Designs - aside from the absence of random assignment, these designs ‘look’ much like a regular experiment: - O1 X O2 - O3 O4 - The only difference is you don’t see the ‘R’s beforehand - Not true experiments (quasi-experiments) - Internal validity issues One Shot Case Study X O - a group is exposed to an event (condition), and then a dependent variable observation is recorded - Example: Hurricane Katrina - the people who live in New Orleans - implement program – to see if it is effective in dealing with the aftermath of Katrina - no comparison group (we don’t know what they would have been like after Katrina with no treatment – we only know with the treatment after Katrina) One Group Pre-test/Post-test Design O1 X O2 - internal validity threats which confound pretest-posttest comparisons - selection – no way of assessing selection in the first place. Static Group Comparison X O1 O2 - selection – it’s an issue – a very big issue - example: cross-sectional survey research o University versus non university X o Income O (income is the observation) o Is exposure to X random? – we cannot make people go to university o Uncontrolled (confounding) variables – personal attributes - If the ‘R’s were here, we would have an experiment Survey Research (Chapter 6 Stuff) - Questions o Behaviour o Attitudes/beliefs/opinions o Sociodemographic characteristics (gender, age, socioeconomic status, etc.) o Intentions o Classifications (classifying people) - Standard Response Options (Variables) o Mutually exclusive and exhaustive (exhaustive means we have category for everybody… which is why researchers put ‘other, please specify’) o Clear and unambiguous Measurement Issues - Validity o External validity (real world) o Internal validity (understand all of these threats to validity) o Face validity (are we measuring what we intend to measure?) - Reliability o Is our measure consistent (reliable) - Understand the different between reliability versus validity o Something can be reliable but not valid – nothing can be valid, but not reliable o If it’s valid, but virtue of being valid – it is reliable  Eg. if he were to stand on a scale – 5 feet tall – reliable (same measurement every time, but it’s not valid) – a scale cannot tell height and he is not 5 feet
More Less

Related notes for SOAN 2120

Log In


Don't have an account?

Join OneClass

Access over 10 million pages of study
documents for 1.3 million courses.

Sign up

Join to view


By registering, I agree to the Terms and Privacy Policies
Already have an account?
Just a few more details

So we can recommend you notes for your school.

Reset Password

Please enter below the email address you registered with and we will send you a link to reset your password.

Add your courses

Get notes from the top students in your class.