Show Menu
Cheatography

Research Design Cheat Sheet (DRAFT) by

Be able to: - Discuss the role of exploration, description, and explanation in social research. - Discuss the logic and procedures of idiographic explanation. - Name and discuss the legitimate and false criteria for nomothetic explanation. - Identify some of the common units of analysis in social research and explain why it is important. - Identify and describe some common study designs based on the time dimension. - Describe the elements and structure of a research proposal

This is a draft cheat sheet. It is a work in progress and is not finished yet.

purposes of emprirical research

1. explor­ation: first orient­ation on new research topic
2. descri­ption: describe observ­ations
3. explan­ation: find out why these observ­ations occur

objects of study

who (or what) is social research about?
- indivi­duals
- groups (e.g. families)
- organi­zations (e.g. faculties, student associ­ations)
- social intera­ctions
- social artifacts

cross sectional + longit­udinal research

cross-­sec­tional research: study based on observ­ations repres­enting a single point in time, a cross-­section of a population
→ make the most condit­ional claims for causal explan­tion, while still able to corrob­ora­te/­falsify hypotheses

longit­udinal: a study design involving the collection of data at different points in time
→ repeated cross-­sec­tional with different people: trend studies
→ repeated with the same set of people; panel studies

compare means

 

Explan­atory research

Explan­atory research implies causal relati­onships between concepts

explan­ations can be:
- idiogr­aphic: complete picture of all causes for observ­ations, using all relevant factors
- nomoth­etic: general unders­tanding of a class of phenomena, using a small number of relevant causal factors
→ the latter contain the hardest criteria for causality

types of explan­atory results
- crosstabs and compare means
- correl­ation
- regression
criteria nomothetic causality:
- a statis­tical correl­ation: changes in one charac­ter­istic are associated with changes in the other
- time order: the cause takes place before the effect
- nonspu­rio­usness
→ Spurious relati­onship: a coinci­dental statis­tical correl­ation between two variables is caused by a
third variable

regression

= the most popular explan­atory statistic
there are different types of regres­sion, but the most common is linear

it is both a biveriate and a multiv­ariate
analysis
*bivar­iate: used when the data set contains two variables and resear­chers aim to undertake compar­isons between the two data set
*multi­var­iate: used when there are more than two variables in the data set.

crosstabs

 

Research Design

research designs make the strongest claims for causal exlana­tion.
time is a relevant dimension

the empiricle cycle consists of the following steps:
1. problem
- research problem: what do you want to know?
- purpose: why do you want to know?

2. theory
- what’s already known based on previous research?
- which concepts are relevant to fix the problem?

3. hypotheses:
- concrete expect­ations based on theory
- deduction: form general theory to testable hypotheses

4. data: how are you going to fix the problem?
- by whom: subjects of study
- when and where?
- how: choice of method data collection and operat­ion­ali­zation?

5. data collection: observ­ations

6. data processing and analyses

7. report results: answer research question and connect to theory (induc­tion)

correl­ation