10 Questions
What is the reason self-administered modes are best fitting for esk.mail/internet?
Uneven diffusion of high-speed internet access combined with low internet proficiency
What do researchers collect to study unobservable variables?
Several related observable variables
What is the purpose of factor analysis?
To determine the relationships between a set of survey items
What is a latent or construct variable?
An unobservable variable
What is the goal of factor analysis in a set of survey questions?
To determine if the questions are measuring the same phenomenon
What is a common application of factor analysis?
To identify relationships between items in a measurement instrument
What is the difference between factor analysis and cluster analysis?
Factor analysis looks at variables, while cluster analysis looks at units
What is the purpose of cluster analysis?
To look for clusters of units that are comparable
What is the input data for factor analysis?
A matrix with variables and units
What does factor analysis help to determine?
If one dimension is being measured or if multiple dimensions are being measured
Study Notes
Nonprobability Polls
- Nonprobability polls have unknown recruiting strategies, making it difficult to determine the probability of a respondent being selected.
- This leads to potential biases, as some individuals may be left out, and the sample may not be representative of the entire population.
- Non-response is another issue, where people do not respond to surveys.
Sample Size
- The sample size depends on the confidence level, confidence interval, or margin of error.
- It also depends on non-response rates.
Survey Research
Positioning Survey Research
- Survey research is deductive, nomothetic, quantitative, positivistic, and pragmatic.
Asking Questions
- Survey research asks questions about behavior, attitudes, beliefs, opinions, characteristics, expectations, self-classification, and knowledge.
- Surveys answer three types of questions: descriptive, relational, and predictive.
Concepts and Variables
- Concepts or constructs are theoretical and abstract.
- Variables are empirical and measured in the world.
- To test hypotheses, theoretical concepts must be translated into empirical variables.
Quantitative Measurement
Levels of Measurement
- There are four levels of measurement: nominal, ordinal, interval, and ratio.
- Scales and indexes are used to measure complex concepts.
- Scales consist of empirically connected parts, while indexes are sums of logically connected parts.
Types of Scales
- Likert scale: statements with categories, scores from 3 to 15, and strong statements.
- Bogardus social distance scale: country, city, neighborhood, and home.
- Semantic differential scale: statements with categories between two ends, such as good to bad.
- Guttman scale: statements with categories, 0=no, 1=yes, and increasing intensity.
Why Do a Survey?
- Surveys are used to collect original data for describing a population too large to observe directly.
- They are used for descriptive, explanatory, and exploratory purposes.
- Scientific surveys are used for explanation, while commercial surveys are used for description.
Steps in Survey Research
- Research design
- Select sample
- Construct questions
- Ask questions
- Analyse data
- Report results
Types of Sampling
- Single stage vs. multistage sampling
- Random vs. systematic sampling
- Stratification: specific characteristics of individuals are represented in the sample.
- Power analysis for sample size determination
- Response rate: the percentage of surveys returned.
Constructing the Questionnaire
- Things to avoid: jargon, slang, abbreviations, vagueness, emotional language, prestige bias, double-barreled questions, beliefs as real, leading questions, and issues beyond respondent capabilities.
- Types of questions: contingency/skip questions and open-ended questions.
Factor Analysis
- Factor analysis is a statistical technique that analyzes relationships between survey items to determine whether they relate to an unobservable variable.
- It is used to determine whether survey items are part of the same scale and measure the same phenomenon.
- There are two ways of analyzing a data matrix: factor analysis and cluster analysis.
Understand the differences between theoretical concepts and empirical variables, and how to measure variables in research. Learn about the levels of measurement and how to apply them.
Make Your Own Quizzes and Flashcards
Convert your notes into interactive study material.
Get started for free
