Unobtrusive Research 1.Content analysis - examine written documents such as editorials. 2.Analyses of existing statistics. 3.Historical/comparative analysis.

Unobtrusive Research

1. Content analysis - examine written documents such as editorials.

2. Analyses of existing statistics.

3. Historical/comparative analysis - historical records.

Comparative & Historical Analysis

• Historically grounded explanations of large-scale and substantively important outcomes

• Sources of Data: Newspapers, Literature Review, Govt. Docs., biographies, diaries, letters, etc.

• Long Tradition: Weber, Durkheim, Marx, Lipset, Skocpol

• Types of Comparative and Historical Analyses– Historical events research: focus on events in

one short historical period– Historical process research: traces a

sequence of events over a number of years– Cross-sectional comparative research:

comparing data from one time period between two or more units.

– Comparative historical research: comparing data from more than one time period in more than one unit

Qualities of Qualitative Historical Research

Case Oriented: focus on unit(s) as a whole

Holistic: how various parts or conditions fit together.

Temporal: taking into account related series of events.

Narrative: researches a story involving actors & events.

Inductive: develops an explanation for what happened

Historical Events Research

Event-Structure Analysis: systematic coding of events/historical information

Griffen (1993) – events leading to the lynching

Oral History: produces written records that can be analyzed

Pagnini & Morgan (1996) – 1170 life histories from the Great Depression

Historical Events Research: an auto manufacturing plant that

produces SUVs closes

Oil embargo

Political crisis

Foreign produced fuel efficient cars increase in popularity

SUV sales decreaseGas

prices increase

Another corporation buys the plant

New owner decreases wages

Workers strike

New owner decides it’s more cost effective to close the plant and move to a less developed country

© Pine Forge Press, an imprint of Sage Publications, 2006

Historical Process Research

• Extends historical events research by focusing on a series of events.

• Can use quantitative data to examine variation over time.

• Example: Number of laws, spending, international agreements, voter turnout, etc.

Comparative Methods

Cross-sectional Comparative Research

Frequently quantitative/variable-oriented research

Comparative Historical Research

Comparisons between cases to highlight the particular features of each case or identify general historical patterns across units.

Paige (1999) - the development of capitalism in various nations.

Cross-Sectional Comparative

Comparative Historical: Lipset (1959)

Comparative & Historical Analysis

Cautions:• Can't trust the accuracy of records - official

or unofficial, primary or secondary.• Must be wary of bias in data sources.• Rarely systematic in data collection

– Based on what’s available– Or on what supports your argument– Measuring across time and contexts

Aviles

• What is the research question?

• What is the theory?

• What is the research design?

• What is the evidence/data?

• What are the findings?

Correlation• MEASURING ASSOCIATION• Establishing a degree of association

between two or more variables gets at the central objective of the scientific enterprise. Scientists spend most of their time figuring out how one thing relates to another and structuring these relationships into explanatory theories. The question of association comes up in normal discourse as well, as in "like father like son“.

Scatterplots

A. scatter diagram

A list of 1,078 pairs of heights would be impossible to grasp. [so we need some method that can examine this data and convert it into a more conceivable format]. One method is plotting the data for the two variables (father's height and son's height) in a graph called a scatter diagram.

B. The Correlation CoefficientThis scatter plot looks like a cloud of points which

visually can give us a nice representation and a gut feeling on the strength of the relationship, and is especially useful for examining outliners or data anomalies, but statistics isn't too fond of simply providing a gut feeling. Statistics is interested in the summary and interpretation of masses of numerical data - so we need to summarize this relationship numerically. How do we do that - yes, with a correlation coefficient.

The correlation coefficient ranges from +1 to -1

r = 1.0

r = .85

r = .42

R = .17

R = - .94

R = - .54

R = - .33

• Computing the Pearson's r correlation coefficient

• Definitional formula is:

Convert each variable to standard units (zscores). The average of the products give the correlation coefficient. But this formula requires you to calculate z-scores for each observation, which means you have to calculate the standard deviation of X and Y before you can get started. For example, look what you have to do for only 5 cases.

Dividing the Sum of ZxZy (2.50) by N (5) get you the correlation coefficient = .50

• The above formula can also be translated into the following – which is a little easier to decipher but is still tedious to use.

yxSSSS

SPr

))(( YYXXSP

2)( XXSSx

2)( YYSS y

• Or in other words …..

22YYXX

YYXXr

• Therefore through some algebraic magic we get the computational formula, which is a bit more manageable.

2222 YNYXNX

YXNXYr

Interpreting correlation coefficients• Strong Association versus Weak

Association: strong: knowing one helps a lot in predicting the other. Weak, information about one variables does not help much in guessing the other. 0 = none; .25 weak; .5 moderate; .75 < strong

• Index of Association• R-squared defined as the proportion of the

variance of one variable accounted for by another variable a.k.a PRE STATISTIC (Proportionate Reduction of Error))

Significance of the correlation

• Null hypothesis?

• Formula:

• Then look to Table C in Appendix B

• Or just look at Table F in Appendix B

21

2

r

Nrt

Limitations of Pearson's r

• 1) at best, one must speak of "strong" and "weak," "some" and "none"-- precisely the vagueness statistical work is meant to cure.

• 2) Assumes Interval level data: Variables measured at different levels require that different statistics be used to test for association.

• 3) Outliers and nonlinearity• The correlation coefficient does not always give a true

indication of the clustering. There are two main exceptional cases: Outliers and nonlinearity.

r = .457 r = .336

4. Assumes a linear relationship

0

10000

20000

30000

40000

50000

60000

0 5 10 15 20 25 30

Education

Sala

ry

Series1

4) Christopher Achen in 1977 argues (and shows empirically) that two correlations can differ because the variance in the samples differ, not because the underlying relationship has changed.

Solution?

Regression analysis