1 / 19

Non-Experimental Research Designs

Non-Experimental Research Designs. Farzin Madjidi, Ed.D. Pepperdine University G.S.E.P. Non-Experimental Designs. Descriptive Relationships Surveys Causal Comparative Qualitative. 2. Research Design looks tough!. But its really easy!. Descriptive Studies.

onslow
Download Presentation

Non-Experimental Research Designs

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. Non-Experimental Research Designs Farzin Madjidi, Ed.D. Pepperdine University G.S.E.P.

  2. Non-Experimental Designs • Descriptive • Relationships • Surveys • Causal Comparative • Qualitative 2

  3. Research Design looks tough!

  4. But its really easy!

  5. Descriptive Studies • Intended to examine and describe an issue • Especially effective when the area has been previously studied • Ex: a study of organizational climate • Ex: a study of salaries in an industry • Typical analysis • Descriptive statistics (graphs, charts, tables, etc.) • Textual analyses (content analysis) 3

  6. Descriptive Studies Important Issues: • Cannot make conclusions about relationships studied • Subjects (AUs) and instrumentation MUST be clearly identified • Watch for “Graphic Distortions” 4

  7. Relationship Studies • Investigate the degree to which differences or variation in one variable are related to differences or variation in another variable • They allow us to examine/discover characteristics that can predict other characteristics • Can lead to further investigation of other variables 5

  8. Relationship Studies • Ex: What is the relationship between leadership styles and tenure as a leader? • Ex: Is there a difference in job satisfaction based on Work Locus of Control? • Typical Analysis • Correlation • Bivariate Regression • Inferential Techniques 6

  9. Predictive Studies • Correlation coefficients only show associations • Predictive studies allow you to calculate the value of one variable (Criterion variable) based on the values of another variable (Predictor variable) • They allow you to make estimates • They allow you to devise forecasting models 7

  10. Predictive Studies • Ex: Determining factors that lead to attrition at high level management positions • Ex: Forecasting property values based on a number of property variables • Typical Analysis includes: • Regression • Multiple Regression • Discriminant Analysis 8

  11. Predictive Studies Important Issues: • Cannot infer Cause-and-Effect • Practical Significance vs. Statistical Significance (low, yet significant correlation; r vs. r-squared) • Reliability and Validity of instruments used • Assumptions/Limitations of statistical techniques used 9

  12. Survey Research • A very popular methodology • Although mostly used in descriptive studies, can be used as part of both descriptive and predictive/relationship studies • Two different types: • Cross-Sectional • Longitudinal 10

  13. Survey ResearchCross-Sectional Surveys • Studies a phenomenon as it occurs at one point in time (e.g., political surveys, attitude surveys, etc.) • Studies a phenomenon through subjects that are along different timelines of the phenomenon (e.g., attitudes of managers in different age groups, surveyed at the same time) 11

  14. 7-98 Cross-Sectional (Mgrs w/ 3 to 5 yrs of exp) Longitudinal 7-96 7-97 7-98 (mgrs w/ 1 yr of exp) (mgrs w/ 2 yr of exp) (mgrs w/ 3 yr of exp) Survey ResearchLongitudinal Surveys • The same group of subjects are studied over a period of time • Avoids some of the limitations of cross-sectional designs Convenience vs. Resources 12

  15. Causal-Comparative Studies Go beyond relationships/associations to examine cause-and-effects. Two types of these studies: • Ex Post Facto • Correlational 13

  16. Causal-Comparative StudiesEx Post Facto • Applied when seeking cause-and-effect relationships, but cannot do experiments • One or more independent variables are used to study their effects on one dependent variable • Ex: What is the impact of a particular training on job performance 14

  17. Causal-Comparative StudiesEx Post Facto (cont’d) Ex. Continued: • Approach 1: Select two groups, one group is trained, the other is not. Next their job performance is measured (experimental design) • Approach 2: Look for groups with pre-existing conditions and compare their job performance. e.g., select a group that is already trained and one that is not trained and compare their job performance. • The groups selected must be as close as possible except for the independent variable 15

  18. Causal-Comparative StudiesCorrelational These studies use more sophisticated versions of correlation analysis to investigate cause-and-effects • Path Analysis:A causal model is developed from theory which shows by arrows the causal sequence that is expected. Correlation between these variables is used as empirical evidence of the proposed links. • Newer, more sophisticated methods: • Structural Equation Modeling • Latent Variable Causal Modeling 16

  19. Causal-Comparative Studies Important Issues: • Primary purpose should be developing cause-and-effect relationships when experimentation is not possible • The “intervention” must have already occurred • Must identify and consider extraneous variables • Differences between the groups not due to the independent variable should be controlled • Be careful with causal conclusions 17

More Related