APA Formatting of tables

Standards for table formatting or the consistent use of abbreviations are becoming increasingly common. Whether for a term paper, dissertation, or publication in a scientific journal, the style and format of a work must meet certain requirements. The standards of the American Psychological Association (APA) are not only widespread in psychology. Here at STATWORX, we also like to use them because APA-formatted tables look visually appealing. But how exactly should my result tables be designed, and can I make the work easier for myself? You'll find answers to these questions here.

Overview of the APA Standard‍

The APA publishes a guide to the publication of scientific articles, known as the Publication Manual, at irregular intervals of several years. It contains guidelines for almost every question on the topic of structuring a scientific paper, writing style, citing, and the clear presentation of empirical results in tables. The sixth edition from 2010 is currently valid and serves as the basis for all the information in this article.

General Guidelines for Tables

When results are to be presented, the first question is whether tables are needed at all. The Publication Manual advises taking a selective approach and considering three aspects. Firstly, it can be very tiring for the reader if many tables need to be worked through. Additionally, it becomes more difficult to mentally follow a text if it is frequently interrupted by tables. Finally, certain information (such as the result of a single variance analysis) can easily be integrated into the text.

The principle that applies here is the same one that generally applies to statistics: Well planned is half done. Therefore, tables should be avoided at all costs. If they truly add value, they should be well integrated into the text. But beware: the Publication Manual clearly states that tables should be self-explanatory. Abbreviations must therefore also be explained there.

Additionally, tables should be numbered in the order in which they are mentioned in the text. The labeling should be Table 1, Table 2, and Table 3, and suffixes like Table 1a and 1b should not be used. Each table must also be referenced in the text. These references should be made clearly, for example, with "as shown in Table 1, ...". References like "see above table" or "the table on page 32" should not be used. Tables in the appendix should also be assigned to the respective section with capital letters. Table A1 would thus be the first table in Appendix A.

Tables According to APA Standards

Let's now take a look at how tables should look in detail according to APA standards. Since pictures say more than words, here is an example of what a simple table according to APA standards looks like.

According to the standards, as few horizontal lines as possible were used, and no vertical lines at all. Clarity was achieved primarily through empty rows and columns between different aspects. The font chosen was Times New Roman with a font size of $12$ pt, which is generally recommended. However, the font should not differ from the rest of the text. Both the table title and the respective abbreviations, as well as the word "Notes", are italicized.

The general notes for the table appear first below the table, including the definition of abbreviations. Individual notes and explanations of abbreviations end with a period and are thus separated from one another. Copyright information also belongs to the general notes and should, of course, always be included.

In a separate line below the general notes, the specific notes follow. These are symbolized by lowercase letters, assigned sequentially from left to right and from top to bottom, starting in the upper-left corner. Multiple specific notes are written consecutively and again separated by a period.

At the very end of the notes, those related to the significance level are listed. These also start in a new line and are each separated by a period. Statements smaller than "*** $p \lt .001$" (not shown in the example) should not be used. If it is necessary to distinguish between one-tailed and two-tailed p-values, the -symbol is used for two-tailed p-values, while a different symbol is used for one-tailed values. For example: " $p \lt .05$, two-tailed. † $p \lt .05$, one-tailed." Since providing an exact p-value is always more informative than simply stating that it is smaller than $.05$ or $.01$, it is recommended to report p-values directly in the table to the second or third decimal place.

In addition to the notes, confidence intervals are also included in the table. Previously, only point estimates were often reported alongside the p-value. However, it is now recommended to also report confidence intervals. The corresponding confidence level should always be explicitly stated (here $95%$), which is essential. Typically, the same confidence level is used for all calculations within an article. However, since each table should be self-contained, the confidence level must be clearly indicated in each case. The chosen format using square brackets in the example is just one possible option. The lower and upper limits of the confidence interval can also be presented in separate columns. This is further illustrated in the following table.

Specialized Results Tables

Depending on the statistical method used, different key figures must be presented to the reader. Below, you will find $3$ different tables, each highlighting a different aspect. This should give you a better idea of how a table should generally be structured according to APA standards. Let's start with a table containing descriptive analyses. The values are presented separately for men and women. A notable feature here is the Cronbach’s $alpha$ value, which is reported without a leading zero, as the value cannot exceed $1$. Additionally, an explanation of the range of values is provided, specifying both the theoretical and the actual range observed in the data. Finally, it is important to mention that the general notes specify the reason for the different sample sizes used in the analysis.

Next, a table is presented that contains all relevant information for a paired-samples t-test. It is crucial that both the mean and the standard deviation are reported for each time point. Additionally, the degrees of freedom for the t-test (here $52$), as well as the confidence interval and the effect size (Cohen’s $d$), are included, as they are essential components of such a summary table.

Finally, we present a table for linear regression, which is one of the most commonly used statistical methods. The table includes $2$ different models, meaning it represents a hierarchical regression. In the first step, only the control variables are included. In the second step, the relevant predictors are added. In addition to the unstandardized regression coefficients, the first model also reports $R^2$ and the corresponding F-value. Together with the sample size stated in the general notes, these values provide sufficient information to evaluate the first model. For the second model, the same information is included in the table, but with additional details: the confidence interval, $\Delta R^2$, and $\Delta F$. The confidence intervals are essential, as they indicate the estimated range in which the true value of the regression coefficients is likely to fall. The $\Delta R^2$ and $\Delta F$ values assess whether the inclusion of the $5$ relevant predictors, in addition to the control variables, significantly explains additional variance in the dependent variable. When all this information is combined, it provides a comprehensive picture of how the predictors relate to the dependent variable.

APA Tables in R

Manually creating APA-compliant tables would certainly be worth the effort for clarity and readability. However, in the age of statistical software, we can (fortunately) save ourselves this work—or at least make it much easier. In R, there is a relatively new package with the straightforward name "apaTables". This package allows users to generate APA-formatted tables in Word with minimal effort. Currently, it supports tables for ANOVA (with and without repeated measures), effect sizes (Cohen’s $\delta$ and $\eta^2$), correlations, and regressions. For regression models, it even allows the generation of bootstrap confidence intervals. As an example, we will now create a correlation table.

# Install and load package
install.packages("apaTables")
library(apaTables)

# Load sample data set
data(attitude)

# Structure of the data set
head(attitude)

# Correlation table according to APA standards
apa.cor.table(attitude, filename = "Korrelationen.doc")

   rating complaints privileges learning raises critical advance
 1     43         51         30       39     61       92      45
 2     63         64         51       54     63       73      47
 3     71         70         68       69     76       86      48
 4     61         63         45       47     54       84      35
 5     81         78         56       66     71       83      47
 6     43         55         49       44     54       49      34

The command apa.cor.table() computes the bivariate Pearson correlation for all variables in the dataset attitude and presents them in an APA-formatted table, including the corresponding confidence intervals. This table is then saved in the file "Korrelation.doc" and contains the following table:

The table already looks very good, but two aspects still need to be manually adjusted.

1. A line break must be inserted before the -notations for p-values, as these notes should be placed in a separate line according to APA standards.
2. All numbers in the table must be centered within their respective cells.

Aside from these minor adjustments, this table—which was generated in just a few seconds—adheres to many APA standards. For example, the abbreviations M for mean and SD for standard deviation are italicized, which is always required when using Latin letters as abbreviations for statistical measures. Additionally, the leading zero before the decimal point is omitted for correlations and p-values, following the APA guideline for numbers that cannot exceed $|1|$.

APA Tables in SPSS

The software SPSS also offers tools to simplify table formatting and directly output APA-compliant tables. To enable this, navigate to SPSS > Settings..., and in the newly opened window, select Pivot Tables. There, you can define the default template for results tables.

The template used here, "APA_TimesRoman_12pt", is not included as a default option in SPSS. It must be manually created, which has the advantage of allowing custom implementation of new APA standards or specific additional requirements.

The template shown in the example can be downloaded here. You then need to place it in the "Looks" folder of SPSS and set it as the default format within SPSS. The "Looks" folder is located in the directory where SPSS is installed. Alternatively, modifying an existing template is well-documented across various online resources.

Summary

Creating APA-compliant tables used to be a time-consuming task, but with modern statistical software, the process has become much more efficient. The effort is well worth it—properly formatted tables not only look professional but are also easier to read and interpret.

If you’d like to learn more about APA standards or need assistance in preparing your statistical results, we at STATWORX are happy to help—feel free to reach out!

References

• American Psychological Association, Publication Manual of the American Psychological Association(6. edition). Washington, DC 2010. ISBN: 978-1-4338-0561-5.