1 / 30

Chapter 7

Chapter 7. Primary Data Collection: Experimentation. What is an Experiment?. Example of a magazine company printing two cover designs and evaluation in the office Example of the same magazine company printing two cover designs and measuring sales in two different cities

sblunt
Télécharger la présentation

Chapter 7

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Chapter 7 Primary Data Collection: Experimentation

  2. What is an Experiment? • Example of a magazine company printing two cover designs and evaluation in the office • Example of the same magazine company printing two cover designs and measuring sales in two different cities • Maker of Grape Jelly trying various formulations

  3. Laboratory Experiment Study in a controlled Situation – outside the natural setting Field Experiment Experiment Study in a realistic Situation – Natural setting

  4. Experiments • Studies in which conditions are controlled so that one or more independent variable can be manipulated to test a hypothesis about a dependent variable. Randomization. • Manipulation of A treatment variable (x), followed by observation of response variable or dependent variable (y). • Goal is to obtain an experimental effect. • Experiment must be designed to control for other variables to establish causal relationship.

  5. Causal relationship is key • Manipulation of variable(s) to observe the effect on another variable • Conditions for causality • Concomitant Variation • Temporal order • Spurious factors • Correlation vs. Causation

  6. Observing an association • If X, then Y • and • If not X, then not Y

  7. Non-spurious • We say that a relationship between two variables is spurious when it is actually due to changes in a third variable, so what appears to be a direct connection is in fact not one. • i.e. If we measure children’s shoe sizes and their academic knowledge, for example, we will find a positive association. • Does that mean that shoe size causes academic knowledge? • What about this? • Do schools with better resources produce better students?

  8. Correlation vs. Causation • Correlation= degree of association between two variable • They must vary together: when one goes up (or down) the other must go up (or down) • Linear relationship • The correlation coefficient can range between +1 and -1. • Positive values indicate a relationship between X and Y variables so that as X increases so does Y. • Negative values mean the relationship between X and Y is such that as values for X increase, values for Y decrease. • A value near zero means that there is a random, nonlinear relationship between the two variables • r- coefficient of correlation

  9. Experimental Setting- Issues • Notation • Design and Treatment • Experimental Effects • Control groups vs. Experimental groups.

  10. Basic Issues • Control Factors • Randomization • Statistical Control • Experimental Validity • Internal Validity • External Validity

  11. Basic Symbols and Notations • Odenotes a formal observation or measurement • Xdenotes exposure of test units participating in the study to the experimental manipulation of treatment • EGdenotes an experimental group of test units that are exposed to the experimental treatment. • CGdenotes a control group of test units participating in the experiment but not exposed to the experimental treatment. • Rdenotes random assignment of test units and experimental treatments to groups. Increases reliability

  12. Experimental Designs One Group, After-only Design EG X O1 Two Group, After-only Design EG X O1 - - - - - - - - - - - - - - CG O2

  13. Experimental Designs (Contd.) • One-group Before-After Design EG O1 X O2 Two-group, Before-after Design EG O1 X O2 • - - - - - - - - - - - - - • CG O3 O4

  14. True-experimental Designs • Two-group After-only Design EG R X O1 • - - - - - - - - - - - - - • CG R O2

  15. True-experimental Designs • Two-group Before-After Design EG R O1 X O2 • - - - - - - - - - - - - - • CG R O3 O4

  16. Internal Validity • The degree to which plausible alternative causes have been controlled for • Are the observed effects on the D.V. a cause of the treatment? Or could they have been caused by something else? Marketing Research Seminar

  17. Internal validity

  18. Threats to Validity • History • Treatment • Maturation • Instrument Variation • Selection Bias • Mortality • Testing Effects • Regression to the Mean

  19. Threats to Internal Validity History: Events external to the experiment that affect responses of the people involved in the experiment (weather, news reports, time of day) -The “cohort effect”: members of one experimental condition experience historical situations different from others Example: Linda McCartney’s death might have affected responses to breast cancer PSAs more for her age cohort; Members of the WW II generation are more responsive to calls for volunteerism and community activism

  20. Threats to Internal Validity Treatment Effect: Awareness of being in the test causes subjects to act different than they otherwise would • Types of treatment effects: • The Hawthorne Effect: special attention received in experiment produces the result • Demand Effect: awareness of test produces response desired by researchers Marketing Research Seminar

  21. Threats to Internal Validity • Maturation: Changes in respondents over the time period of the experiment (maturing, getting hungry, getting tired) Marketing Research Seminar

  22. Threats to Internal Validity • Testing Effect: A before treatment measurement sensitizes subjects to the treatment • Example: Colon Cancer PSA (phoning subjects for pre-test measurements may have sensitized subjects to ads that appeared on TV) Marketing Research Seminar

  23. Threats to Internal Validity • Instrumentation Effects: The measuring instrument may change, • different interviewers may be used, or • an interviewer or confederate gets tired • A common case: order of presentation produces an effect • Example: consumers may prefer first product tasted if they can’t tell the difference Marketing Research Seminar

  24. Threats to Internal Validity • Mortality (or attrition): Some subjects drop out of the experiment between measurements. • Those subjects who drop out may differ from those who stay • Example: testing a weight-loss program Marketing Research Seminar

  25. Threats to Internal Validity • Selection Bias: An experimental group is different from control groups • For convenience, many experimental studies have self-selected subjects • random assignment to treatments will solve this • Example: Latin students Marketing Research Seminar

  26. External validity

  27. Experimentation: Pros and Cons • Best method to evaluate causation • Costs • Security • Implementation Issues

  28. Steps for starting a good design • 1. Select problem • 2. Determining dependent variables • 3. Determining independent variables • 4. Determining the number of levels of independent variables • 5. Determining the possible combinations • 6. Determining the number of observations • 8. Randomization

More Related