Research Methodology summary

Chapters 1-3

Methodological approaches to the social world:
1. topic must concern social phenomena
- researcher subject involves people (how they act, think, feel and interact)
- looks at the groups that people form, relations within and among such groups, how they adapt to changes in
society
2. it must be possible to address the topic or answer the question by making appropriate observations

Scientific questions/theories:
- can be answered by identifying the conditions under which observable events take place
- requires concepts that describe the phenomena of interest, general laws and theories that identify patterned
relationships
- events to be explained can be logically deduced from laws and theories (to explain past, present and future)
- shouldn’t be judged as true or false, only useful

Empiricism – the only admissible evidence is that which we can know (in)directly through our senses

Objectivity – the capability of independent observers to agree on the results of observations

Control bias and errors in the course of the research

Approaches to social research:
1. Experiment
- Often offer the best approach for investigating the causes of phenomena; for specific conditions
- The researcher is in control of the treatment
- Outcome variable (consequence or result) is measured after the manipulation
- Manipulation of the “cause” (=test condition, independent variable or treatment)
- Use comparable groups (groups should be as similar as possible, only differing in test condition)

2. Survey research
- Use of questionnaires or interviews; when can’t impose an outcome with a treatment
- Large or representative samples
- Efficient method of data collection
- Purpose – describe the frequency of certain characteristics among groups

3. Field research
- Direct observation of the behavior – if you want to see how people react, you need to observe them
- Gain firsthand knowledge of the situation
- Immersing oneself in a naturally occurring set of events

4. Available data
- Data that have been generated for purposes other than those of the researcher
- Written records, newspapers, documents
- Non-verbal physical evidence like paintings
- Non-reactive measurement
- Data doesn‘t need to be „created“, but still needs to be „collected“
- Problems: how to codify the data in a systematic fashion; what to do when the information is insufficient/not
comparable
Secondary data analysis = analyzing data that are made available by other researchers




Opportunities -> which main approach is most suited?

,  The opportunity to observe individuals in their natural living environment
 The opportunity to generalize the research findings
 The opportunity to replicate the research
 The opportunity to formally test if a social event or fact has an effect on behavior and opinion of people
 The opportunity to research new and relatively unknown social problems

Difficulties -> which main approach has the most difficulty with…?

 Access to data or data collection
 Problems related to evaluating the quality of measurement (validity/reliability)
 Risks regarding researcher (interviewer) bias
 Risks regarding subject (test person, respondent) bias

Theory <- proposition -> hypothesis
- Proposition = general statement regarding a regularity in the behavior or opinion of subjects
e.g., “when an individual manages a particular task well, then he will perform that task better in the presence of others
than when nobody else is present”

- Theory = provides an explanation for a proposition or set of propositions, NOT speculation
e.g., “nature (something biological) vs. nurture (you’re vulnerable to the fact that you’re being observed) explanations for
this proposition
‘biological’: the presence of others activates physiological triggers
‘psychological’: people perform better when they believe they are being watched/evaluated

- Hypothesis = applying the proposition in a concrete (real life) situation
e.g., “athletes perform better when there are a lot of spectators watching them performing than when there is hardly any
public”




Deduction - the conclusion is absolutely certain if the evidence is true/ to (start with the hypothesis) valid or invalid
Induction - the conclusion is uncertain even if the evidence is true because the content goes beyond the evidence (end up
with the hypothesis) bottom up; vary in strength or probability

Concepts (constructs) = general/abstract description of a social phenomena
e.g., ethnocentrism

Variable = empirical manifestation of a social phenomenon; characteristics of units that may vary in successive observations
Quantitative variables have categories that express numerical distinctions
Qualitative variables involve differences in kind rather than in number
e.g., a scale that measures ethnocentrism

Hypothesis = an expected relationship between 2 (or more) variables that can be researched/tested
e.g., women are on average less ethnocentric than men
Total effect hypothesis – an overall effect of independent and dependent variable

, Type of hypothesis and the conceptual model
 Bivariate hypothesis: expected relationship between 2 variables
Diagram:
X ---- Y
X: independent variable (cause) – from which a certain arrow is departed, but not arriving there
Y: dependent variable (outcome) – arrows only arrive, don’t depart
->: direction of effect (from independent on dependent)
Mediator – both sends and receives

 Multivariate hypothesis: expected relationship between more than 2 variables (between a dependent variable Y and
multiple independent variables X..)

o Multiple causality – the relative importance of independent variables

++ = X1 has a bigger influence on Y than X2 (positive effect)
- = if X increases, then Y decreases (vice versa)

Example:

“The likelihood of re-entering in the job market increases
with educational level and decreases with the level of
unemployment benefit. The effect of education is stronger
than the effect of the unemployment benefit”
o Mediation = the effect of X on Y is indirect
through its mediating variable (X2)
= X1 -> X2 -> Y
X3 has an indirect effect on Y, not expecting a direct
correlation between X3 and Y

e.g. the more opportunities to enter the labor market, the
higher the likelihood of re-entering




Partial mediation = direct + indirect effect (a loop/total effect)




o A moderating effect = the effect of X1 on Y is conditional on the moderator X2 (depending on the value of X2
= conditional effect (intensifier (+) or supperssor (-) effect)




Example:
“The higher the willingness to work, the higher the likelihood of
re-entering the labor market. This effect is intensified by the
level of opportunity to re-enter.”