Dependent variable DV or Y Effect or Outcome variable measures the effect of

# Dependent variable dv or y effect or outcome variable

This preview shows page 10 - 15 out of 44 pages.

Dependent variable (DV, or Y :( · Effect” or Outcome variable – measures the effect of the independent variables on the test units, e.g., sales, profits and market shares Extraneous variables : · All variables other than the independent variables that affect the response of the test units, this might provide a false causal effect
How to control Randomization or Random Assignment · Random assignment of test units to experimental groups by using random numbers · Treatment conditions are also randomly assigned to experimental groups Matching · Comparing test units on a set of key background variables before assigning them to the treatment conditions Statistical control · Measuring the extraneous variables and controlling for their effects through statistical analysis Design control · Use of experiments designed to control specific extraneous variables
Pre-experimental design X: IV / exposure of a group to a treatment O: DV / the process of observation or measurement of outcome variable R: The random assignment of test units or groups to separate treatment conditions EG : Experimental group CG : Control group One-shot case study X O1 · Single measurement after treatment · No experimental or control groups, or randomization · No control for extraneous variables Example: Effectiveness of Bunnings TV advertising Telephone interviews to people who watched the commercial (X) and then willingness to purchase (O) are measured · Possible extraneous variables: history, selection bias One Group Pretest-Posttest Design O1 X O2 · Measurements before and after treatment · No experimental or control groups, or randomisation
· No control for extraneous variables · Treatment Effect is O2 - O1 Example: Effectiveness of Bunnings TV advertising Get respondents in a room. Ask to rate Bunnings (O1), watch the commercial (X) and then rate Bunnings afterwards (O2 ( · Possible extraneous variables: history, selection bias Static design EG: X O1 CG: O2 · A two-group experimental design · The experimental group (EG) is exposed to the treatment, and the control group (CG) is not · Measurements on both groups are made only after the treatment · Test units are not assigned at random · The treatment effect is O1 – O2 Example: Effectiveness of Bunnings TV advertising Telephone interviews – ask people if they watched the ad (X). If yes, they are in EG, else in CG. Ask each group to rate Bunnings · Possible extraneous variables – Selection bias True experimental design ** Employ randomization to control history & selection bias
Posttest-only control group design (A&B TESTING ( EG: R X O 1 CG: R O 2 · Similar to static group design, except that test units are randomly assigned to experimental or control group · The treatment effect is obtained by TE = O 1 - O 2 Selection bias (SB) is eliminated by randomization · But cannot be checked · Timing & Cost: 2 groups, 1 measurement per group – simple to implement and cheap Pretest-posttest control group design EG: R O 1 X O 2 CG: R O 3 O 4 · Similar to Posttest-only control group design, except that measurements before and

#### You've reached the end of your free preview.

Want to read all 44 pages?

• Two '16