12 ©2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. Experimental Research Research Methods in Management
©2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. What is experiment? 2
©2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. 12–3 Creating an Experiment Experiment An experiment is “an orderly procedure carried out with the goal of verifying, falsifying, or establishing the validity of a hypothesis”. An experiment provides insight into cause-and-effect by demonstrating what outcome occurs when a particular factor is manipulated.
©2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. 12–4 Experimental treatment: the way an experimental variable is manipulated. Manipulation of the Independent Variable X X Y Y Experimental treatment is applied Change is observed Independent variableDependent variable
©2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. Types of experimental design Pure experimental design Quasi experimental design
©2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. 12–6 Subjects The sampling units for an experiment, usually human respondents who provide measures based on the experimental manipulation. Experimental Group A group of subjects to whom an experimental treatment is administered. Control Group A group of subjects to whom no experimental treatment is administered. Experimental and Control groups
©2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. 12–7 Does giving more assignments improve student performance? Experimental and Control groups Assignment Student Performance Amount of assignmentsTest score Operationalization
©2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. 12–8 Experimental subjects Experimental Group: A lot of assignments Control Group: Regular load of assignments Experimental and Control groups
©2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. The Effect of the treatment Experimental Group Control Group Average Performance (E) = 90 Average Performance (E) = 90 Average Performance (C) = 80 Average Performance (C) = 80 The Effect of the experimental treatment = (E - C) = = 10 A Treatment (a lot of assignments) are given
©2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. Pretest-Posttest Experimental Group Control Group Average Performance (E2) = 90 Average Performance (E2) = 90 Average Performance (C2) = 75 Average Performance (C2) = 75 Average Performance (E1) = 70 Average Performance (E1) = 70 Average Performance (C1) = 70 Average Performance (C1) = 70 Before Giving A Treatment (Pretest) After giving A Treatment (Posttest) The Effect of the treatment = (E2 – E1) – (C2 – C1) = (90 – 70) – (75 – 70) = 20 – 5 = 15 A Treatment (a lot of assignments) Is given
©2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. Within-Subjects Designs Between-Subjects Designs Measure The Effect of the treatment
©2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. 12–12 Involves repeated measures because with each treatment the same subject is measured. Example: Will salary increase improve job performance? Within-Subjects Designs Salary $1,000 Average job performance Sold 10 units/month
©2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. Within-Subject design 13 Increased salary to $1,100 Increased salary to $1,200 Increased salary to $1,300 Sold 15 units/month Sold 20 units/month Sold 25 units/month
©2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. Within-Subject Design Longitudinal study Researchers conduct several observations of the same subjects over a period of time
©2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. 12–15 Each subject receives only one treatment combination. Usually advantageous although they are usually more costly. Validity is usually higher. Between-Subjects Designs
©2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. Between-Subjects Design 16 Increased salary to $1,100 Increased salary to $1,200 No salary increase Measured job performance Sold 15 units/month Measured job performance Sold 25 units/month Measured job performance Sold 10 units/month !!!!!
©2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. Between-Subject Design Cross-sectional study Researchers compare different population groups at a single point in time
©2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. 12–18 Quasi-experimental design Quasi-experiment has all the same elements as pure experiment; however, subjects are not randomly assigned to conditions. Subjects are selected based on the values of the independent variable (e.g., age, gender), rather than having the experimenter assign values of the independent variable to subjects
©2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. Quasi-experimental design Example: The effect of Amphetamine on work performance. Subjects who are prescribed Amphetamine by a doctor will be assigned to the experimental group. Control group Experimental group
©2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. 12–20 Issues in Experimental Research Internal Validity The extent that an experimental variable is truly responsible for any variance in the dependent variable. ◗ Does the experimental manipulation truly cause changes in the specific outcome of interest? The placebo effect The measurable, observable, or felt improvement in health or behavior not attributable to a medication or invasive treatment that has been administered.
©2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. 12–21 Issues in Experimental Research Hawthorne Effect People will perform differently from normal when they know that someone observes them
©2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. 12–22 Issues in Experimental Research External Validity The accuracy with which experimental results can be generalized beyond the experimental subjects. e.g. can the results from a small group experiment be generalized to a larger group of people?