09_Lecture_for_PSYC2018H_F24

Chapter 9: Non-Experimental Design I: Survey Methods

Explanation of the significance of sampling in survey research compared to other psychology research.
Principles of good survey construction to ensure data validity.
Problems encountered in interpreting survey data.
Four methods of collecting survey data along with their advantages and disadvantages.

Simple Random Sampling: Every individual has an equal chance of being selected. Used to generalize findings to a larger population.
Stratified Sampling: Population divided into strata (groups) and random samples taken from each stratum. Useful for ensuring representation across major characteristics.
Cluster Sampling: Population divided into clusters (often geographically) and entire clusters are randomly selected. Cost-effective and efficient for large populations.

Positive Correlation: Both variables increase together (illustrated through scatterplots).
Negative Correlation: One variable increases as the other decreases.
Coefficient of Determination (r²): Measurement of the proportion of variability in one variable that can be explained by the other.

Simple Linear Regression: Predicting scores using a single predictor variable. Formula: Y = a + bX (where Y is the predicted value, X is the predictor).
Multiple Regression: Predicts scores using multiple variables, exploring the interaction between predictors.

Directionality Problem: Difficulty determining which variable is the cause and which is the effect.
Cross-Lagged Panel Study: Helps ascertain the temporal order of variables, addressing directionality issues.
Third Variable Problem: Other variables might contribute to the observed relationship. Evaluated through partial correlation.

Bias vs. Representation: Importance of ensuring representative samples to avoid biases that can affect results and interpretations.
Non-Probability vs. Probability Sampling: Differences and implications for survey outcomes.
Self-Selection Bias: Historical example of Literary Digest's failed election predictions due to sampling bias (subscribers + ownership of phones).

Open-ended Questions: Allow detailed responses, but are harder to analyze.
Closed Questions: Easier to analyze, but can limit responses.
Likert Scales: Useful for gauging attitudes but may lead to response biases.
Demographic Information: Should be included at the end of surveys to avoid influencing responses.

Wording and Clarity: Important to avoid ambiguity and leading questions.
Pilot Study: Helps identify potential issues in survey questions before widespread deployment.

Pros: Cost-effective, efficient.
Cons: Must be brief; risk of SUGging (selling under the guise of surveying).

Correlation techniques allow researchers to find relationships between variables without inferring causation.
Scatterplots: Visual representation of data relationships.

Regression Analysis: Helps in making predictions about one variable based on another; essential for understanding predictor effects on outcomes.

Mediators vs. Moderators: Distinction between variables that explain how and why relationships exist versus those that highlight the conditions under which relationships exist.
Importance of thorough analysis of variable interactions when interpreting results.

Surveys are essential tools for gathering data on attitudes, beliefs, and projected behaviors.
Careful survey design and analysis techniques such as correlation and regression are vital for understanding data.
Researchers must remain vigilant about directionality challenges and third variable influences throughout their analysis of non-experimental data.