L2: conceptualization and variables

Operationalization

Process of turning abstract concepts into measurable variables and indicators

what is the process for operationalization

1. Convert theoretical terms into variables

2. Determine indicators for the variables

3. Determine characteristics of the indicators

Variables Vs indicators

Variables: are properties or characteristics of the concept (e.g., socio-economic status)

indicators: are ways of measuring or quantifying variables (e.g., income)

What concepts do we need to operationalize here: Is there a relationship between the social status of parents and the educational success of their children?

• social status of parents

• Educational success

What is a unit?

What we collect data on

Ex. Individuals, countries, companies, years,

*units can be combined → individuals in different countries over several years (nested data)

what is a variable?

Characteristics that can vary between units

Ex. For dogs, fur color, weight, breed

what are variable characteristics?

• Name: what is the variable called? (education, var_46, edurec2)

• Values: possible data points of a variable (1, 2, 3, 99, 999)

• Value labels: translation of numeric value to meaning
  (1 = "unemployed",', 2 = "retired"...)

• Missing: How are lack of data values labeled? (999 = missing)

• Measurement level: NOIR (Nominal, Ordinal, Interval, Ratio)

What does the measurement level NOIR stand for?

• nominal

• Ordinal

• Interval

• Ratio

NOIR: N

Nominal: data can only be categorized

Eg: fur color (white, brown, black)

NOIR: O

Ordinal: Data can be categorized and ranked, not evenly spaced

Eg. Beginner, intermediate, expert

NOIR: I

Interval: data can be categorized and ranked, evenly spaced

No true zero

Eg. Year of birth

NOIR: R

Ratio: data can be categorized, ranked, evenly spaced, and has a natural 0

Eg. Years in school (1-10)

Level of measurement: categorical variable

nominal

Ordinal

level of measurement: continuous variable

interval

Ratio

Hypotheses

• a tentative, provisional, or unconfirmed statement derived from theory that can be either verified or falsified.

• It explicitly states the expected relationship between an independent and dependent variable that we can test

  • "The higher the X, the higher the Y"

• A large part of quantitative empirical research consists of hypothesis testing

What is a population?

• = entire group the researcher wants to study

• Example: All eligible voters in Sweden, students at Gothenburg university

• Problem: often too large to be surveyed

What is a sample?

• = subset of the population that actually participates in the survey

• Key in survey research

Population and sample

Representative sample

• Reflects the characteristics of the population

• Allows researchers to generalize findings to the population

• Example
  Population: 52% women, 48% men
  → Sample should have a similar composition.

Random sample

Often basically impossible to match the population characteristics in the sample

→ random sample

• Every member of the population has an equal chance of being selected

• Random sampling increases the likelihood of obtaining a representative sample

Key idea:

Random # representative, but random sampling is the best way to achieve representativeness

Biased sample

• Some groups are overrepresented or underrepresented

• Problem: Results may not reflect the true population -> cannot generalize

Two common sources of bias in surveys:

Selection bias:

• Some people have a higher chance of being included than others.

• Example: online survey advertised on Instagram excludes people who do not use Instagram

Non-response bias

• Some selected people are less likely to participate.

• Example: survey about politics is mainly completed by politically interested citizens

Summary

• Operationalization is the process of turning abstract concepts into measurable indicators

• Variables are properties of characteristics of the concepts that vary between units

• Units are the things that we collect data on (individual, countries, etc.)

• Variables have different level of measuresments Nominal, Ordinal,
  Interval, Ratio (NOIR)

• A hypothesis is a statement about the relation between variables that can either be verified or falsified