Week 9: Quantitative Analysis I

What is quantifying data?

The numerical representation and manipulation of observations to describe and explain phenomena

Give an example of quantifying data.

Coding individuals’ responses numerically to study the relationship between job insecurity and health

What is the quantification process?

Converting raw data into a numerical format for computer-based analysis

What is a codebook?

A document that outlines variable names, numerical codes, value labels, and missing value codes

How might gender be coded in quantifying data?

1 = Male

How can religious affiliation be coded?

For example, 1 = Roman Catholic

How is job insecurity coded?

1 = “Not at all likely”

What is the purpose of data cleaning?

To detect and correct coding errors in datasets

What is a key task in data cleaning regarding codes?

Verifying that only the defined codes appear in the data file

How should out-of-range responses be handled?

They should be corrected or recoded to a predefined value (e.g., 98 for “Don’t know”)

What is contingency cleaning?

Checking for consistency across related variables (e.g., ensuring a respondent's gender remains consistent)

What is univariate analysis?

The examination of the distribution of cases for one variable

What are frequency distributions?

Counts that show how many times each attribute occurs, typically presented in tables or charts

How are frequency distributions commonly displayed?

In tables, bar charts, or pie charts (for nominal/ordinal variables with fewer than 5 categories)

How is proportion calculated?

p = (f/n), where f is frequency and n is the total number of cases

How is percentage calculated?

Percentage = (f/n) x 100

What are measures of central tendency?

The mean, median, and mode

What is the mean?

The arithmetic average of values

What is the median?

The middle value in a ranked set of observations

What is the mode?

The most frequently occurring value

How can skewed distributions affect measures of central tendency?

The mean and median can differ significantly in skewed distributions

What are measures of dispersion?

The range and standard deviation

What does the range measure?

The difference between the highest and lowest values

How is standard deviation calculated?

By computing the mean, subtracting the mean from each observation, squaring the deviations, summing them, dividing by the number of observations, and taking the square root

What does a large standard deviation indicate?

That there are many outliers and the data is spread out

What does a small standard deviation indicate?

That the data is clustered closely around the mean

What is bivariate analysis?

The analysis of the relationship between two variables

What technique is used in bivariate analysis to present relationships?

Cross-tabulations (contingency tables)

How can the relationship between gender and religious attendance be examined?

By separating data by gender, classifying frequency of weekly versus less-than-weekly attendance, and calculating percentages

What is multivariate analysis?

The simultaneous analysis of relationships among three or more variables

What is the purpose of multivariate analysis?

To explain a dependent variable using multiple independent variables

Give an example of multivariate analysis.

Investigating the effect of age, gender, and social class on religiosity

Why is a codebook important?

It serves as a roadmap for data analysis by clearly documenting every variable and its coding

How do outliers affect statistical measures?

They can disproportionately affect the mean while the median and mode remain more robust

Which type of chart is preferred for ordinal data?

Bar charts

When are pie charts typically used?

For nominal data with fewer than 5 categories

Why are data consistency checks critical during data cleaning?

They maintain the integrity of the dataset by ensuring responses are recorded consistently