5.3 Describing Continuous Variables
represent quantities or amounts. Examples include age, reaction time, total score, average scale score, or number of correct responses.
For continuous variables, we usually describe:
- sample size and missing values
- center
- spread
- range
- shape
Measures of Center
The most common measures of center are the , , and .
For many continuous variables, the mean is the most commonly reported measure of center. However, the median can be especially useful when a variable is skewed or has outliers.
Measures of Spread
The most common measures of spread include:
- minimum and maximum
- quartiles and
The standard deviation is especially common when reporting means. The IQR is especially useful when reporting medians.
Measures of Shape
Descriptive output can also help you understand the shape of a distribution. Two useful values are and .
Skew tells you whether one tail of the distribution is longer than the other. Positive skew means the tail extends toward higher values. Negative skew means the tail extends toward lower values.
Kurtosis is related to tail behavior. You do not need to obsess over kurtosis at this point, but it can help flag distributions that are unusually heavy-tailed or light-tailed.
Running Continuous Descriptives in jamovi
To describe continuous variables in jamovi:
- Go to Analyses.
- Select Exploration.
- Select Descriptives.
- Move your continuous variable or variables into the Variables box.
- Under Statistics, select the descriptive statistics that match your purpose.
For many early analyses, useful options include:
NMissingMeanStd. deviationMinimumMaximumSkewnessKurtosis
You do not need to select everything jamovi can produce. More output is not automatically better.
Interpreting Continuous Descriptives
When looking at continuous descriptives, ask:
- What is a typical value?
- How much do values vary?
- What are the minimum and maximum values?
- Are there missing values?
- Does the variable appear skewed?
- Do the values fall within the expected range?
If the descriptive statistics look strange, return to your dataset and check your variable setup, transformations, and computed variables.
A mean scale score is based on items rated from 1 to 5. The descriptive statistics show a minimum of 0.25 and a maximum of 6.10. What should you do?
Answer
You should check your data preparation and formula. A mean score based on items rated from 1 to 5 should not be below 1 or above 5.