SOCI 217 Lecture Notes - Lecture 13: Western Electric, Blind Experiment, Regression Analysis
Week 6- experiments
C. experimental and control groups
• Hawthorne effect
o Term coined in reference to a series of productivity studies at the Hawthorne
plant of the western electric company in Chicago, Illinois.
o If people know they are being studied, people change their behaviour
D. double blind experiment
• Experimental design in which neither the subjects nor the experimenters know which is
the experimental group and which is the control group
• Researchers giving placebo do not know which group is the experimental or control
group
Selective subjects
• Experimental and control groups should be as similar as possible
• 1. Probability sampling
• 2. Randomization
o randomly assign half to experimental and the other half to control
• 3. Matching
o pairs of subjects matched based on their similarities
o then one is assigned to experimental and other to control group
• 4. Matching or randomization
o randomization with large samples
o matching for small samples
variation on experimental design
• pre-experimental research designs
• 1. One shot case study
o single group of subjects measured on dependent variable following the
administration of some experimental stimulus
o guy exercises and then check weight
• 2. One group pretest-protest design
o check weight then exercise then check weight again
• 3. Static group comparison
o some research based on experimental and control groups but no pretests
find more resources at oneclass.com
find more resources at oneclass.com
Document Summary
Week 6- experiments: experimental and control groups, hawthorne effect, term coined in reference to a series of productivity studies at the hawthorne plant of the western electric company in chicago, illinois. Selective subjects: experimental and control groups should be as similar as possible, 1. Randomization: randomly assign half to experimental and the other half to control, 3. Matching: pairs of subjects matched based on their similarities, then one is assigned to experimental and other to control group, 4. Matching or randomization: randomization with large samples, matching for small samples variation on experimental design, pre-experimental research designs, 1. One shot case study: single group of subjects measured on dependent variable following the administration of some experimental stimulus, guy exercises and then check weight, 2. One group pretest-protest design: check weight then exercise then check weight again, 3. Static group comparison: some research based on experimental and control groups but no pretests. Sources of internal invalidity: history, time, maturation testing, hawthorne effect.