Class Notes (809,049)
Canada (493,506)
Sociology (2,049)
Lecture 17

2Z03 Lecture 17 "Experiments".docx

6 Pages
Unlock Document

McMaster University
Gerald Bierling

Fox 1 Lecture 17 SOCIOL 2Z02 Monday March 24, 2014 Experiments:  Lecture outline: • Topics Appropriate to Experiments • The classical experiment • Selecting subjects • Threats to validity • Strengths and weaknesses of experiments • Example • Natural experiments  Experiments involve doing 2 things: • Taking action (eg. showing a film) • Observing results of that action (eg. does opinion or behavior change?) o Experiments are not used as often as other data collection methods: • Many of the issues we study are beyond our ability to manipulate (eg. economic inequality) • In social sciences, used more often in fields such as psychology and economics • Can sometimes take advantage of natural experiments  Topics Appropriate for Experiments: • Hypothesis testing (eg. does awareness increase recycling behavior?) trying to isolate a cause • Better suited for explanatory (another way of thinking about hypothesis testing) research than descriptive research (due to sample size limitations) • Opinion formation (why these are formed), small group interaction (how do people behave in certain conditions in small groups?)  The Classical Experiment: • 3 Major Components: 1) Independent (cause) and dependant (effect) variables: o IV (stimulus)  DV (outcome) Fox 2 Lecture 17 2) Pre-test & post-test: o Measure DV (recycling behavior)  Introduce stimulus (film)  Re-measure DV o Pre-test: the measurement of a DV among subjects (just at different points in time) o Post-test: the measurement of a DV among subjects, after they have been exposed to an IV o Examples of experiment include reading comprehension levels in schools o If there is a difference in variables, we can begin to talk about causality (dependant variable will take on a certain value) • 2 Problems (at this point in assuming causality (X causes Y) o Internal validity: did something else have an influence on the change in the DV? (between post-test and pre-test) (Ex= students are naturally getting better at reading comprehension as they go throughout the school year=>not comprehension program=internal validity) o External validity: Can results be generalized? Are subjects representative? • Deal with validity problems in 2 ways: i. Sampling/subject selection: o How to sample people that participate in the study and assign them to a specific group ii. Use of experimental & control groups: o Want 2 groups in our experiment 3) Experimental & Control Groups: • Experimental group: receives stimulus (show them film about recycling) • Control Group: doesn’t receive stimulus (receive placebo sugar pill); should resemble the experimental group in all other respects (should look exactly alike as possible) Experimental Group Control Group Measure Attitudes (DV) Are attitudes the same? Measure attitudes (DV) Watch film (IV) (stimulus) ------forget for awhile-------- Re-measure attitudes (DV) are atttudes the same? Re-measure attitudes (DV) • If attitudes are the same=film has no effect Fox 3 Lecture 17 • If attitudes in experimental group changed more than control group, gives some bases for showing a film or making them aware of things, will impact their attitudes and behavior • If attitudes of experimental group become more positive than control group, then you can still say the film has made an impact (IV caused a certain outcome) • A control group helps deal with validity issues- eg. Hawethorne effect (subjects knew they were being studied, and by having 2 groups, subjects don’t know who is being studied) • Importance of subjects not knowing the group to which they are assigned • Further strengthened by the use of a double-blind approach: subjects or people don’t know where subjects belong to (people taking measurements don’t know because they can record results in a bias manner) • Must be equal in every way except not knowing they are being studied  Selecting subjects: • Probability sampling: not possible due to limited number of subj
More Less

Related notes for SOCIOL 2Z03

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.