Week 10 Textbook Ch. 14: Quantitative Data Analysis

Quantitative Data Analysis

  • Introduction to Quantitative Data

    • Quantitative data analysis is prevalent in sociology.
    • It can be intimidating for students due to math anxiety.
    • Understanding the intuition behind statistical methods is crucial.

  • Importance of Numbers in Media

    • Examples show the prevalence of quantitative data in news articles:
    • Increase in white supremacist propaganda by 120% from 2018 to 2019 (Schor 2020).
    • Comparison of medical worries among racial groups (Murphy 2020).
    • Hydroxychloroquine's effects in a study of Covid patients (Nigam and Cohen 2020).

  • SPSS for Data Analysis

    • SPSS (Statistical Package for the Social Sciences) facilitates data processing.
    • Accessible to students in various institutions.
    • The chapter focuses on practical applications rather than mathematical calculations.

Preparing for Data Analysis

  • Key Learning Objectives

    1. Understand data entry and the role of a codebook.
    2. Find existing datasets for secondary data analysis.
    3. Describe the format of datasets in SPSS.

  • Data Entry Process

    • All data must be structured into a dataset:
    • A structured collection typically stored in computer files.
    • Use codebooks for guidance in the entries.

  • Role of a Codebook

    • A codebook outlines the content, layout, and coding of data.
    • Coding transforms survey responses into numerical data for analysis.
    • Numeric coding simplifies data entry and reduces errors.

  • Types of Variables

    • Dichotomous variables (e.g., yes/no questions) represented as dummy variables (1 = yes, 0 = no).
    • Categorizations may vary in complexity (nominal, ordinal, scale).

Handling Missing Data

  • Missing data should be categorized distinctively in analysis.
  • Non-responses to sensitive topics may be common and should be marked.
  • Coding missing responses with a unique identifier (e.g., 999 for non-response) helps in managing data accuracy.

Conducting Univariate Data Analysis

  • Univariate Analysis

    • Analysis of one variable yields descriptive statistics.
    • Frequency distributions provide insights into sample characteristics.

  • Measures of Central Tendency

    • Mode: Most frequent value in the distribution.
    • Median: Midpoint value, dividing the data into two equal halves.
    • Mean: Average calculated by summing values and dividing by the count.

Measures of Variability

  • Central tendency alone doesn't give insights into variability in data.
    • Range: Difference between lowest and highest values.
    • Interquartile Range (IQR): Measures distance between 25th and 75th percentiles.
    • Standard Deviation (SD): Average distance of all scores from the mean.

Bivariate Data Analysis

  • Crosstabulations

    • Examines relationships between two categorical variables, identifying correlations.
    • Crosstabulation tables show how variables intersect and can suggest deeper insights or differences.

  • The Chi-Square Test

    • Helps to determine if observed frequencies of two variables significantly differ.
    • Evaluates whether relationships observed in sample data can be generalized to the population.
    • Rejects the null hypothesis if p-value < .05 (indicating statistical significance).

  • Interpreting Crosstab Results

    • Analyze patterns, strengths, and directions of relationships between variables using percentage point differences.
    • Emphasize significant findings and their implications for broader societal conclusions.

  • Conclusion

    • Both univariate and bivariate analyses are foundational for understanding and interpreting sociological data.
    • Quantitative methods provide valuable frameworks for data analysis that inform social theories.