CMNS 260 Lecture Notes - Lecture 7: Okcupid, Latin Square, Interrupted Time Series
1
Experiments & Coding
Experiments
• Often used in applied research to test reactions of audiences in media studies or to test
products.
Rise in popularity of experimental research in the social sciences (early & mid-20th c.)
• Behaviorism (e.g. B. F. Skinner)
• Quantification (e.g. I.Q. tests – Alfred Binet)
• Applications: scientific management (F. W. Taylor)
• Eaple: Colour oditioig to ifluee peoples ehaiour
o Baker-Miller Pik aka Druk Tak Pik used to al iolet prisoers i US prisos
Historical Trends in Experimental Social Research
• Types of subjects (changes in contemporary ethics)
o Self or named individuals (19th century)
o Captie populatios historial i 9th ad earl th , ot o
▪ Institutionalized groups, like prisoners, psychiatric patients, etc.
▪ Fator orkers, et. Hathor effet
o Volunteers (paid, unpaid) & consenting participants (21st c.)
• Reporting style
o Naming vs. anonymity
• Deception & risk
o Contemporary research ethics guidelines for treatment of human subjects: potential for
benefits must outweigh risk of harm
• Debriefing (Including offers of professional help)
General Types of Experiments
• Laboratory experiment: Controlled experiment in artificial environment (example: Zimbardo
prison experiment.
• Field experiment: The execution of an experimental design in real-world conditions. P.196
• Natural experiment: An experimental approximation in which group assignment and
administration of the independent variable is outside the researcher's control (i.e., occur
naturally). These are often considered quasi-experimental designs due to the lack of control.
The language of experiments (Keywords)
• Experimental research is research in which the researcher intervenes or does something to one
group of people but not to another, then compares results for the two groups
• Experimental group: a group of subjects to whom an experimental stimulus is administered.
• Control group: a group of subjects to whom no experimental stimulus is administered and who
should resemble the experimental group in all other respects. The comparison of the control
group and the experimental group at the end of the experiment points to the effect of the
experimental stimulus. P.181
find more resources at oneclass.com
find more resources at oneclass.com
2
Key Terms
• Treatment (design notation: X): a term used for an independent variable in experimental
research.
• Pretest (design notation: O1): measurement of the dependent variable of an experiment prior to
treatment (before exposure to independent variable).
• Post-test (design notation: O2): measurement of the dependent variable in experimental
research after the treatment (exposure to independent variable).
• Dichotomous variable: A variable that has only two attributes; also called a binomial variable.
Design Notation
• O = observation
• X = treatment
• R = random assignment
The language of experiments: Selecting Participants
• Random assignment is a practice that involves dividing subjects into groups at the beginning of
experimental research using a random process, so the experimenter can treat the groups as
equivalent. (design notation: R)
o Randomization: The procedure of randomly assigning experimental subjects to
experimental and control groups. P.184
o Matching: The procedure whereby pairs of subjects are matched on the basis of their
similarities on one or more variables; then one member of the pair is assigned to the
experimental group and the other to the control group. P.184
Recall: Stanford Prison Experiment
• Used random assignment to check equivalence of participants (tested for psychological
characteristics)
• Assigned subjects who seemed equivalent to play either prisoners or guards
• Research question: do institutions like prisons cause people to behave in specific ways?
The language of experiments: Experimental Design
• The classical experiment: conventional type of experiment involves three major pairs of
components:
1. Independent and dependent variables,
2. Pre- testing and post-testing, and
3. Experimental and control groups. P.179
• The double-blind experiment is a type of experimental research in which neither the subjects
nor the person who directly deals with the subjects for the experimenter knows the specifics of
the experiment. They do not know which is the experimental group and which is the control.
Recall: Milgrim obedience study
• Single blind design
o Sujets ere teahers ad did ot ko that learers ere researhers
find more resources at oneclass.com
find more resources at oneclass.com
Document Summary
Experiments: often used in applied research to test reactions of audiences in media studies or to test products. Historical trends in experimental social research: types of subjects (changes in contemporary ethics, self or named individuals (19th century) (cid:858)capti(cid:448)e(cid:859) populatio(cid:374)s (cid:894)histori(cid:272)al i(cid:374) (cid:1005)9th a(cid:374)d earl(cid:455) (cid:1006)(cid:1004)th (cid:272), (cid:374)ot (cid:374)o(cid:449)(cid:895) Laboratory experiment: controlled experiment in artificial environment (example: zimbardo prison experiment: field experiment: the execution of an experimental design in real-world conditions. P. 196: natural experiment: an experimental approximation in which group assignment and administration of the independent variable is outside the researcher"s control (i. e. , occur (cid:858)naturally(cid:859)). These are often considered quasi-experimental designs due to the lack of control. The comparison of the control group and the experimental group at the end of the experiment points to the effect of the experimental stimulus. Design notation: o = observation, x = treatment, r = random assignment.