How to Create a Stem-and-Leaf Plot in SPSS

Data Preparation and Example Scenario in SPSS

Before initiating the procedure, it is vital to ensure your numerical data is correctly structured in the SPSS Data View. A Stem-and-Leaf Plot is specifically designed for quantitative measurements. The dataset used here serves as a practical illustration, showing the average points per game for sixteen distinct basketball players. Each observation must be correctly entered under a defined numerical variable in the datasheet.

The plot displays data by splitting up each value in the dataset into a stem and a leaf. This is a highly useful plot for easily visualizing the underlying distribution of a dataset while preserving the exact numerical scores. For researchers, this visual check is often the first step in assessing normality and identifying potential data irregularities.

Suppose we have the following dataset that shows the average points per game for 16 basketball players, ready for analysis in the SPSS Data Editor:

The variable is named ‘points’ and contains integer values ranging from 5 to 31. This range suggests that the stem will primarily represent the tens digit, and the leaf will represent the ones digit, providing an efficient summary of the central tendencies and spread of the athletes’ performance scores.

Navigating to the Explore Dialogue Box

To generate a Stem-and-Leaf Plot for this dataset, you must utilize the comprehensive data exploration features found under the Descriptive Statistics menu. This pathway ensures that the plot is generated alongside crucial numerical summaries, providing a holistic overview of the variable’s characteristics. The process begins by selecting the primary analytical tab.

Specifically, click the Analyze tab located in the top menu bar of the SPSS window. This action opens the main statistical procedures menu. From there, select Descriptive Statistics. This module is dedicated to summarizing data and examining distributions. Finally, select the Explore option from the submenu. The ‘Explore’ command is robust, designed to perform detailed univariate analysis, including the generation of specialized plots like the Stem-and-Leaf display.

The systematic click sequence is: Analyze → Descriptive Statistics → Explore. This sequence is necessary because the Stem-and-Leaf Plot is integrated into the exploratory analysis toolkit, rather than being located under the dedicated Graphs menu.

This action will successfully launch the “Explore” dialog box, which is the control center for defining the parameters of your analysis.

Configuring Variables and Plot Options

The “Explore” dialog box presents various options for structuring the analysis. Our primary goal here is to specify which variable to plot and to ensure that the graphical output is enabled. This requires correctly moving the variable of interest into the designated list and confirming the output display setting.

This will bring up the following window:

To create the Stem-and-Leaf Plot, we need to move the variable **points** from the list on the left into the box labelled **Dependent List**. Variables placed in this list are the ones on which the descriptive and graphical analyses will be performed. This identifies ‘points’ as the quantitative variable whose distribution we seek to visualize.

Next, you must confirm the setting under the option that says **Display** near the bottom of the box. Ensure that **Plots** is selected. Alternatively, selecting **Both** will provide the complete numerical summary tables alongside the required graphical outputs. If only “Statistics” is selected, the Stem-and-Leaf Plot will not be generated, regardless of other settings.

After configuring these settings, click the **OK** button to execute the command and generate the output in the SPSS Viewer.

Generating and Interpreting the Plot Output

Once we click OK, the SPSS Output Viewer will immediately display the results, including the various tables of descriptive statistics and the visualizations. The resulting Stem-and-Leaf Plot will appear shortly after the initial summary tables.

The structure of the plot is divided into two primary columns: the **Stem** column displays the first digit (or leading digits) for each data value, while the **Leaf** column displays the second digit (or trailing digit). This arrangement allows for direct reconstruction of the original numerical observations.

For example, the first leaf shown in the first row corresponds to the stem ‘0’. Combined, the stem (0) and the leaf (5) represent the observation of 5 points. This observation indicates the lowest average score in the sample and highlights the presence of a potential low-end outlier relative to the main body of scores clustered in the tens and twenties.

Conversely, by examining the final row, we identify the highest score. The last leaf shown in the last row (leaf 1) corresponds to the stem ‘3’. This reconstruction indicates the player who averages 31 points per game—the maximum score in the dataset.

This simple graphical representation successfully helps us get an immediate visual idea of the distribution of the points scored by the 16 players in this dataset, confirming that the majority of scores fall within the 10-29 range.

Advanced Analysis of Distribution Shape

The primary power of the Stem-and-Leaf Plot lies in its use as a diagnostic tool for assessing the shape of the numerical variable’s distribution. By visually rotating the plot, you can evaluate characteristics such as symmetry, unimodality, and the presence of extreme scores with precision that histograms often lack due to binning effects. The goal is often to determine if the data approximates a normal distribution, a key assumption for many inferential statistical tests.

In the basketball points example, the plot shows a noticeable concentration of leaves around stems 1 and 2. This clustering indicates the **central tendency** of the data. If the distribution were perfectly normal, the leaves would form a bell-shaped curve. While our plot is reasonably symmetric, the single observation at stem 0 (5 points) and the observation at stem 3 (31 points) define the **spread** and range of the data, alerting the researcher to the endpoints of the performance scale.

Analysts should also look for signs of **skewness** (asymmetry) or **kurtosis** (peakedness). If most leaves pile up on the lower stems with a long tail extending to the higher stems, the distribution is positively skewed. Conversely, if leaves cluster at the high end, it is negatively skewed. By integrating this visual assessment with the numerical measures provided in the accompanying Descriptive Statistics tables, researchers gain a comprehensive understanding of the variable characteristics before proceeding to subsequent model building.

Conclusion on Using Stem-and-Leaf Plots in SPSS

The procedure for creating a Stem-and-Leaf Plot in SPSS is straightforward, utilizing the dedicated Explore command under Descriptive Statistics. This method is highly efficient for data screening, particularly for smaller datasets where retaining the exact numerical value of each observation is an advantage over traditional histograms.

By correctly identifying the numerical variable and ensuring the **Plots** display option is activated, users of SPSS can leverage this technique to gain immediate, meaningful insights into their data’s distribution. The resulting plot provides clear visual evidence of data patterning, clustering, and the presence of outliers, which are essential considerations for validating statistical assumptions.

Mastering this technique ensures that your initial data exploration phase is robust and minimizes the risk of basing subsequent inferential analyses on data that violate critical distributional assumptions. Always use the combined visual and numerical output provided by the Explore function for the most reliable assessment of your data.