Research Methods Sociology

Scientific Research

The Controlled, objective, systematic search for knowledge

• has the goal of eliminating the error and effects of extraneous variables

Variable

A characteristic that varies from one person/phenomena to another

Quantitative Variables

Discrete (finite/infinite)

• something that can be described with a number (# of cars, GPA)

Qualitative

Continuous (has intervals)

• Can be described using words

Purposes of Research

To be…

1. Descriptive - describe what something is

2. Predictive - predict a phenomena that will occur in the future

3. Improve - study the effectiveness of interventions

4. Explain - explain causality of phenomenon

Research Design

1. Specification of the Problem

2. Literature review (conceptualize, theoretical grounds)

3. Hypothesis/research questions

4. Operationalize

   1. Defining ways to measure variables

5. Data Collection method

6. Sampling

7. Data Analysis

8. Report the Result

Quantitative Study

Descriptive and inferential statistics

• Generalization is the goal

Qualitative Study

Exploratory and use of words

• in-depth studies

  • development of themes (major concepts)

Population

A total set of objects that you study

• Parameters - characteristics of this

Sample

A subset of the population

• Statistics - characteristics of these

Quantitative Data Analysis

Concerned with numbers

• statistics = quantitative data analysis

There are two types…

Descriptive and Inferential

Deductive Approach

Hypothesis testing (quantitative)

• Topic → theory → hypothesis → data collection → hypothesis testing

Inductive Approach

Theory development (qualitative)

• Topic → data collection → data analysis → empirical generalization/explanation

Null

No relationship at all, concepts operate independently of each other

• ice cream sales and shark attacks

Covariance

Concepts vary together, but one is not the cause and the other the effect, can be negative or positive

• salary and price of milk

Causal

Concepts co-vary (are related) and changes in one concept precede changes in the other

• diet and weight gain

Measurement

is the assignment of numbers to objects, messages, or events according to certain rules

• Assigned numbers are observations

Nominal Measurement

Includes the least information

• Numbers or symbols used to simply classify an object or event

There is assumed equivalence with categories

• Blue eyes, brown eyes, green eyes, grey eyes

• Type of medium (TV, newspaper, radio)

Ordinal Measurement

Involves rank-ordering objects on some dimensions

• there is no assumption that there are equal intervals between scale points

Interval Measurement

Rank ordering and the intervals are equal

• This scale includes an arbitrary zero point, meaning zero does not indicate an absence

Ratio Measurement

All characteristics of interval scale, but has a true zero point (zero means absence)

• length/height. # of something. Kelvin Temperature (0 = no heat at all)

if you can say that one object possesses X times as much of a property as another, it is this

Implications levels of measurement

Analysis requires minimum levels of measurement

• some variables can be treated as multiple levels of measurement as well

Measurement Reliability

measuring something in a consistent and stable manner

• Test-retest method - researcher administers the measuring instrument to the same group of people at 2 different times and calculated correlations to get a reliability estimate

  • Making the same measurement more than once

Measurement Validity

How well researchers are measuring what they intend to measure

• Content - measuring instrument covers all attributes of the concept you are trying to measure, nothing is left out

• Construct - when findings of measurement are meaningful in more than a descriptive sense

• Face - investigator’s subjective evaluation of the appropriateness of the instrument for measuring the concept rather than whether the instrument measures what the researcher wants to measure

• Sampling - whether a given population is adequately sampled by the measuring instrument in question