Practical Issues in Social Research Methods

6,954 views
6,529 views

Published on

Practical issues in social research methods.

Published in: Technology, Business
0 Comments
0 Likes
Statistics
Notes
  • Be the first to comment

  • Be the first to like this

No Downloads
Views
Total views
6,954
On SlideShare
0
From Embeds
0
Number of Embeds
24
Actions
Shares
0
Downloads
54
Comments
0
Likes
0
Embeds 0
No embeds

No notes for slide

Practical Issues in Social Research Methods

  1. 1. Lecture January 11, 2010
  2. 3. Formulation of Theoretical Model and Research Problem (1) <ul><li>THERE SHOULD BE THE POSSIBILITY OF SURPRISE IN SOCIAL RESEARCH. RESEARCH PROBLEM AS A PUZZLE </li></ul><ul><li>Selecting Research Question </li></ul><ul><ul><li>A difference between advocacy research and scientific research </li></ul></ul><ul><ul><li>(Advocacy research refers to research that sifts through evidence to argue a predetermined position) </li></ul></ul><ul><ul><li>(Scientific research does not suppress contrary or inconvenient evidence) </li></ul></ul>
  3. 4. Formulation of Theoretical Model and Research Problem (2) <ul><li>Researchable question </li></ul><ul><ul><li>What to avoid? </li></ul></ul><ul><ul><ul><li>Questions that imply answers dealing with different moral or aesthetic values </li></ul></ul></ul><ul><ul><ul><li>Questions that answering them involves unethical procedure </li></ul></ul></ul><ul><ul><li>Good questions </li></ul></ul><ul><ul><ul><li>What proceeds why? Galileo’s maxim: description first, explanation second </li></ul></ul></ul>
  4. 5. Formulation of Theoretical Model and Research Problem (3) <ul><li>Interesting question </li></ul><ul><li>“ The heart of good work is a puzzle and an idea” (Abbott 2003, p. xi). </li></ul><ul><li>The no-surprise objection : “the answer is already well documented”, “we know answer before we do research” “the question is trivial” </li></ul><ul><li>The “so what” objection : “no relevance for social theory or for social life” </li></ul>
  5. 6. Formulation of Theoretical Model and Research Problem (4) <ul><li>Good research questions </li></ul><ul><ul><li>Proposing new research </li></ul></ul><ul><ul><li>Challenging prior research </li></ul></ul><ul><ul><li>Extending prior research </li></ul></ul><ul><ul><li>To formulate good research problem and a theoretical model requires an extensive review of the literature. </li></ul></ul><ul><ul><li>Looking for a good review articles that provide (1) a theoretical grid or template, (2) an overview of key findings and unresolved issues, and (3) a description of the most influential studies. </li></ul></ul>
  6. 7. Formulation of Theoretical Model and Research Problem (5) <ul><li>Choosing variables and specifying hypotheses </li></ul><ul><li>At minimum, any hypothesis involves two variables: an independent variable and a dependent variable. </li></ul><ul><li>- “You can’t explain a variable with a constant.” Maximizing variance to find the effect of a cause </li></ul><ul><li>- Substantive profiling: The use of telling comparisons </li></ul>
  7. 8. Preparation of Research Design (1) <ul><li>A research design is a plan that shows, through a discussion of the model and data, how we expect to use our evidence to make inferences. </li></ul><ul><li>Model implies variables, units, and observations (values). </li></ul><ul><li>Data collection refers to observation, participant observation, intensive interviews, large-scale surveys, histories recoded from secondary data, ethnographies, randomized experiments, and other types. </li></ul>
  8. 9. Preparation of Research Design (2) <ul><li>How the data are collected? </li></ul><ul><li>Decisions: What data are available? What additional data will be needed? </li></ul><ul><li>Data collection is costly in terms of money and time. </li></ul><ul><li>- Sources of funding </li></ul><ul><li>- Timetable </li></ul><ul><li>We have to know how the data will be used. </li></ul>
  9. 10. Preparation of Research Design (3) <ul><li>We have to know how the data will be used. Discussion of data analyses methods </li></ul><ul><li>Multi-method approaches </li></ul>
  10. 11. Measurement (1) <ul><li>Types of sources: </li></ul><ul><li>Verbal reports (self-report). Surveys </li></ul><ul><li>Observation (firsthand, or through various devices) </li></ul><ul><li>Archival records (statistical documents such as censuses, diaries, mass communications, and others) </li></ul><ul><li>Criteria of good measurement: </li></ul><ul><li>Valid </li></ul><ul><li>Reliable </li></ul><ul><li>Exhaustive </li></ul><ul><li>Mutually Exclusive </li></ul><ul><li>All involve measurement errors </li></ul>
  11. 12. Measurement (2) <ul><li>Observed reality = True reality + Error </li></ul><ul><li>Error = True reality - Observed reality </li></ul><ul><li>Minimizing errors through multi-indicator approach </li></ul>
  12. 13. Measurement (3) <ul><li>Missing data </li></ul><ul><li>Traditional approach </li></ul><ul><li>New approach: </li></ul><ul><li>- regression imputation </li></ul><ul><li>- random assignment </li></ul>
  13. 14. Sampling (1) <ul><li>. </li></ul>
  14. 15. Sampling (2) <ul><li>Target population --|> Frame population: Coverage error </li></ul><ul><li>Frame population --|> Selected sample: Sampling error </li></ul><ul><li>Selected sample --|> Collected sample: Non-response error </li></ul><ul><li>Coverage error and non-response error as the most serious errors in both qualitative and quantitative research </li></ul>
  15. 16. Sampling (3) <ul><li>Special issues: </li></ul><ul><li>Samples for focus groups </li></ul><ul><li>Samples for Internet studies </li></ul>
  16. 17. Data collection (1) <ul><li>Politics of data collection </li></ul><ul><li>Data collection as a social process. </li></ul><ul><li>Sociology of data collection: Who needs what data for what purpose? </li></ul>
  17. 18. Data collection (2) <ul><li>Quality control of data collection </li></ul>
  18. 19. Analyses and interpretation (1) <ul><li>Statistics and substance in causal inferences </li></ul><ul><li>Where the logic of qualitative and quantitative research is the same and – where it is different? </li></ul>
  19. 20. Analyses and interpretation (2) <ul><li>Special issues of causal inferences: </li></ul><ul><li>- endogeneity </li></ul><ul><li>- types of errors </li></ul><ul><ul><li>Type I (α): reject the null hypothesis when the null hypothesis is true, and </li></ul></ul><ul><ul><li>Type II (β): accept the null hypothesis when the null hypothesis is false </li></ul></ul>
  20. 21. Seven rules (1) <ul><li>Glenn Firebaugh, Seven Rules for Social Research. Princeton: Princeton University Press, 2008 </li></ul><ul><li>1. THERE SHOULD BE THE POSSIBILITY OF SURPRISE IN SOCIAL RESEARCH. RESEARCH PROBLEM AS A PUZZLE </li></ul>
  21. 22. Seven rules (2) <ul><li>2. LOOK FOR DIFFERENCES THAT MAKE A DIFFERENCE, AND REPORT THEM </li></ul><ul><li>3 . BUILD REALITY CHECKS INTO YOUR RESEARCH </li></ul><ul><li>4 . REPLICATE WHERE POSSIBLE </li></ul>
  22. 23. Seven rules (3) <ul><li>5. COMPARE LIKE WITH LIKE </li></ul><ul><li>6 . USE PANEL DATA TO STUDY INDIVIDUAL CHANGE AND REPEATED CROSS-SECTIONAL DATA TO STUDY SOCIAL CHANGE </li></ul><ul><li>7 . LET METHOD BE THE SERVANT, NOT THE MASTER </li></ul>

×