There are a few go to techniques that we use for survey data analysis. The primary tool used for analyzing survey data is the crosstab. There are fancier multivariate techniques, and those have their place, but for everyday use the crosstab is the preferred method for analyzing nominal and ordinal data. Questions that generate these data types dominate most consumer and B2B market research surveys.
The crosstab (xtab for short) can accommodate two, or more, variables. It’s purpose is to examine the shared distributions of the variables. When used with a statistic, such as the chi-square, it is used to assess the degree of association between variables. Please note that I did not say causation, but association, this important distinction is reserved for another discussion.
The sample below provides an illustration of a two variable xtab with three levels for the row variable and two for the column variable (3 x 2). A third variable can be added to ‘control’ for potential influence. In this case we could add gender to see if the relationship between current job security and the desire to look for a new job is impacted by gender.
Our Job Security * Look for Another Job Crosstabulation
Look for another job YesNoTotalOur Job SecuritySecureCount73421552889% within Our Job Security25.40%74.60%100.00%% within Look for Another Job19.90%36.00%29.80%SomewhatCount131428864200% within Our Job Security31.30%68.70%100.00%% within Look for Another Job35.60%48.20%43.40%InsecureCount16.439.5125.94% within Our Job Security63.30%36.70%100.00%% within Look for Another Job44.50%15.90%26.80%TotalCount36.9159.9296.83% within Our Job Security38.10%61.90%100.00%% within Look for Another Job100.00%100.00%100.00%Chi-square tests
Value df Asymp. Sig. (2-sided)Pearson Chi-Square
980.497* 2 .000 Likelihood Ratio 968.016 2 .000 Linear-by-Lindear Association 807.954 1 .000 Not Valid Cases 9683 * 0 cells (0%) have expected count less than 5.The minimum expected count is 988.79.
We view crosstabs from the perspective of rows and columns. For example we can say that 25% of those who feel secure in their current position are indeed looking for a new position. This compares to 63% of those who have an insecure feeling about their current position. In other words, those who feel insecure about their current position are over twice as likely as those who feel secure to be looking for a new position. From the column perspective 45% of those who are looking for a new position feel insecure about their current employment. This compares to 16% for those who are not looking for a new job.
When making comparisons we can compare one cell to either another cell or to the row or column totals. For example, 31% of those who feel only somewhat secure in their position are looking for new work, compared to 63% for those who feel insecure and 38% overall.
In short, we can say that a person’s desire to seek new employment is associated with their sense of security in their current position. This type of information would be very useful to corporate HR managers tasked with employee satisfaction and retention.
Crosstabs are extremely useful in analyzing relationships between two or more nominal or ordinal variables. They are the primary analytical tool for market research. So take a few minutes and get to know your rows and columns.
You might find these other articles on crosstab and causation helpful: