(Week 4)Research design and measurement:

Question 1

What is research design:

Answer 1

Clearly specify what you want to find out
Determine the best way to do it
Answer questions: why, who and what, when, where, and how

Question 2

The three ‘Why’ / Three purposes of research:

Answer 2

1. Exploratory: Familiarizing/Satsify curiosity/ Basis for future study/ Qualitative
2. Descriptive: Reports on characteristics/ What are the main features/ (Who, what, when, where)/Qualitative
3. Explanatory: To make plain/ (How and Why) / Quantitative

Question 3

Who vs What

Answer 3

Who = the objects of our investigation/things we want to make claims about
—Unit of Analysis- the “thing” whose characteristics we are seeking to describe or explain/The specific object to which evidence (data) refer
——————————————————————–
What = what is being measured, what we collect data from
Units of Observation
The kinds of objects from which evidence is collected
Can be the same as the units of analysis, but not necessarily

Question 4

What are the types of unit analysis:

Answer 4

Individuals(Most typical, any type of individual)
Groups(Social groups/specific traits of social groups)
Organizations(Formal social organizations)
Social Artifacts (Any product of human activity)

Question 5

Two common fallacies in the Who/What domain:

Answer 5

Ecological fallacy: The reasoning error that occurs when conclusions about individuals are based solely on group observations(For aggregate data, evidence about the unit of analysis informs us only about the unit, not about the components of the unit)

Exception Fallacy: The reasoning error that occurs when conclusions about aggregates are drawn from individual cases
(Many think that if their individual experience differs from the aggregate pattern, then the aggregate pattern is negated
Identifying an exceptional case does not dismiss an aggregate finding)

Question 6

Two types of study in the When Dimension

Answer 6

Cross-sectional: A study based on observations representing a single point in time - A snapshot
Explanatory and descriptive studies are often cross-sectional
Con: Difficult to generalize from one-point-in-time observations

Longitudinal: Study design involving the collection of data at different points in time
At least twice: two waves
Pro: Accounts for change over time
Con: Costly, complex design; people may drop out

Question 7

Three types of Longitudinal studies:

Answer 7

1. Trend: a given characteristic of a general population is monitored over time/Different people at multiple points in time/Detect trends or patterns
2. Pannel: data is collected from the same set of people (the sample or panel) at several points in time.
   Offers the most comprehensive data on changes over time, comparing before and after an event
   ——Panel Attrition: participants drop out of the study, which reduces the accuracy of measuring changes
3. Cohort: A specific subpopulation or cohort is studied over time, although data may be collected from different members in each set of observations

Question 8

Conceptualization and Operalization:

Answer 8

Conceptualization: To develop a conceptual definition, explain what a concept means in the research context, and what is and is not being studied. It is different for each of us.
Concepts have:
Indicators - the observation we choose and consider to reflect a concept
Dimensions - aspects of the concept we are interested in observing

Operalization: The development of specific research procedures (operations) that will result in empirical observations representing those concepts
Specifies the meaning of a concept by showing how it is experienced empirically

Question 9

All variables have to have two characteristics

Answer 9

Exhaustive: Cover the whole range of a variable

Mutually exclusive: A response cannot be represented by more than one attribute

Question 10

Level of measurements:

Answer 10

Nominal: Lowest level/just discrete categories/ cannot be ranked bu can be classified)/ Same or different/ Qualitative
——If they are different or the same equal or not equal

Ordinal: Can logical rank/same different and more or less/no indication of the magnitude of difference between attributes/ Qualitative
———-If one is more than the other , > or <

Interval: Attributes are ranked order and have equal distance between adjacent attributes/ Same or different/more or less/ and how much difference/
———How much they differ + or -

Ratio: Variable whose attributes have all the qualities of nominal, ordinal, and interval measures, and in addition are based on a “true zero” point/The same or different & more or less/higher or lower amounts & how much more different & ratios to describe differences (Quantitative)
——–The ratio of one to the other , x or /