Which Statistical Analysis Should You Use?
Selecting a Statistical Analysis to Answer Your Research Question
Figuring out which type of statistical analysis to use to answer your research question is a crucial step but it can also be confusing and frustrating. That’s why we created this tool—to help you quickly and confidently find an analysis that is right for you and your research needs.
There are many types of statistical tests available. But most research questions can be answered with just a handful of options. If your research question is similar to the majority of those asked, this tool can help identify a statistical analysis that will work for you.
It will walk you through a series of questions and based on your answers, recommend a “best fit” analysis, all while explaining the steps and options.*
Are you ready to get started?
*The current iteration of this tool presumes you are most likely working with parametric data. I.e., with normal rather than non-normal distributions. However, some of the most popular non-parametric correlations are provided.
What Are You Testing For?
Your answer to this first question is extremely important. Start by thinking about exactly what it is you are trying to find out. Are you identifying differences between two groups, examining the relationships between two variables, or trying to predict relationships between variables and/or group inclusion?
- Identifying differences in group means
- If you want to see whether the difference between the arithmetic means (average value) of something in one or more groups is statistically significant compared to the same in other groups, you are looking for differences in group means. A statistically significant difference is one that is unlikely to have occurred by chance. An example of this type of study is one comparing the average (mean) size of oak tree leaves in one location to the average size of oak tree leaves in other locations.
- Examining relationships between two variables
- If you want to determine whether and how variable values affect each other, you are examining relationships between variables. If your research question sounds like any of the following, in which X is one variable and Y another, you are looking for the relationship between two variables. Is X related to Y? What is the relationship between X and Y? How are X1 and X2 related to Y?
- Predicting relationships and/or group inclusion
- These are research studies where you are trying to predict a dependent variable’s outcome value from one or more other variables, or predict group inclusion in two or more groups from one set of variables.
How Many Dependent Variables Do You Have?
Dependent variables (often denoted by Y) are variables whose values depend on other variables. In your study, dependent variables are whatever you plan to measure to see if they change or vary as a result of changes in the other (independent) variables. For example, if you are comparing the average size of oak leaves in one location (group) to the average size of oak leaves in another location, the dependent variable is the average oak leaf size at each location.
Because you are testing differences in group means, your dependent variables will be type Interval. Tell Me More
How Many Independent Variables Do You Have?
Independent variables are variables (often denoted by X) whose values do not depend on the other variables. In your study, they are the values that you plan to manipulate or change. The change may also be a natural occurring difference, such as gender or eye color. In a study comparing the average size of oak leaves in one location to the average size of oak leaves in another location, the independent variables are the locations.
Because you are testing differences in group means, your dependent variables will be type Categorical. Tell Me More
Do You Have a Covariate?
A covariate is a variable that is possibly predictive of or influences the outcome of the statistical study. A covariate may be of direct interest (e.g., the influence of a participant’s age) or it may be a confounding variable. A confounding variable is an extraneous variable in an experimental design that correlates with both the dependent and independent variables.
Think of covariates as any variable that operates alongside the variables that are the focus of your analysis. They can confuse or confound the relationship you are investigating, so you have to control for them to ensure your results are accurate. Covariates are sometimes referred to as intervening, extraneous, controlled or hidden variables, all terms that provide clues about what they are and how to handle them. For example, if you are studying the effects of geographic location and gender on height—whether and how height is influenced by these two variables—a covariate might be participants’ age at measurement.
Covariates are usually measured at the onset of the study, so that you can account for their effects during your statistical analysis.
What Types of Variables Are You Examining?
When testing the relationships between two variables, both variables may be the same type or they may differ. Which of the following types of variables will you be examining in your research?
- Ordinal
- An ordinal variable is one where the order of the variable’s values (categories) matter but the difference between those values does not matter. For example, in a race someone finishes first, another person finishes second, and a third person comes in third. There is a specific order to the individual placements, but the interval measures between the finishing times are unimportant.
- Interval
- For an interval variable, the intervals between the values of the variable are equally spaced. The difference between those values do matter because they are meaningful.
- Continuous
- These are variables that can have an infinite number of different values between two given points.
- Nominal
- Also known as a categorical variable, this is a variable with values that do not have any implied ordering or intrinsic numeric value. Examples: political parties, religion. Important note: If your nominal (categorical) variable has only two categories, choose the Dichotomous option.
- Dichotomous
- A Dichotomous variable has only two values (categories), such as male or female.
Keep in mind that you are examining the simple relationships between two and only two variables. When examining such bivariate (two-variable) relationships, neither variable is considered a dependent variable. Rather, both are considered independent variables and your task is to determine the nature of the relationship between the two. For example, you may chose to study the relationship between ice cream sales and time of day.
What is Your Core Requirement?
To answer this question, think about how your variables relationship is structured. Based on the variable types you identified previously, the following core requirement options are available to you.
Because both of your variables are type Interval, your core requirement is Linear Normal. This is a linear (approximately straight line) relationship between two variables. For example, a frequency distribution of a variable that looks like a normal curve (a bell shape).
Because both of your variables are type Continuous, your core requirement is Linear Normal. This is a linear (approximately straight line) relationship between two variables. For example, a frequency distribution of a variable that looks like a normal curve (a bell shape).
Because one of your variables is ordinal and one interval, your core requirement is True Ranking. You will determine the degree to which two rankled variables agree with each other with regards to their ranking.
Because one of your variables is continuous and the other dichotomous, your core requirement is Only Two Choices. A dichotomous variable has only two possibilities or choices.
How many Predictor Variables do you have?
In analyses which involve prediction there are always two identifiable variables. One of those variables (and most often there is more than just one) is called the predictor variable. Predictor Variables are independent variables that individually or in conjunction with other predictor variables predict an outcome (dependent) variable.
What type of Outcome Variable(s) do you have?
The other variable that is involved in prediction is called an outcome variable. Outcome Variables are the variables that are being predicted in analyses. In order to select the correct test, we'll need to know the type of your outcome variable(s).
Which Statistical Analysis Should You Use?
Recommended for You! Based on what you are researching and the number and types of variables you plan to include, you could use a: One-Way Anova or T-Test
Do you want to know more about this type of analysis?
Capella’s Quantitative Skills Center on iGuide provides helpful statistical resources and support, such as Choosing the Correct Statistical Test. This hour-long presentation describes various types of statistical analyses and offers detailed information and guidance.
The Web also has many resources that can help you understand, structure and conduct your statistical analysis. Try searching for them using the name of the recommended statistical analysis and the types of variables you will use as your search terms. Here is an example of the resources you might find using those search terms:
- One-Way Analysis of Variance (ANOVA) Example Problem
- Using SPSS for One Way Analysis of Variance
- Hypothesis Testing: T-Tests
- SPSS Annotated Output: T-Test
Do you want to start over and get a different recommendation?
Which Statistical Analysis Should You Use?
Recommended for You! Based on what you are researching and the number and types of variables you plan to include, you could use a: One-Way Anova
Do you want to know more about this type of analysis?
Capella’s Quantitative Skills Center on iGuide provides helpful statistical resources and support, such as Choosing the Correct Statistical Test. This hour-long presentation describes various types of statistical analyses and offers detailed information and guidance.
The Web also has many resources that can help you understand, structure and conduct your statistical analysis. Try searching for them using the name of the recommended statistical analysis and the types of variables you will use as your search terms. Here is an example of the resources you might find using those search terms:
Do you want to start over and get a different recommendation?
Which Statistical Analysis Should You Use?
Recommended for You! Based on what you are researching and the number and types of variables you plan to include, you could use a: Factorial Anova
Do you want to know more about this type of analysis?
Capella’s Quantitative Skills Center on iGuide provides helpful statistical resources and support, such as Choosing the Correct Statistical Test. This hour-long presentation describes various types of statistical analyses and offers detailed information and guidance.
The Web also has many resources that can help you understand, structure and conduct your statistical analysis. Try searching for them using the name of the recommended statistical analysis and the types of variables you will use as your search terms. Here is an example of the resources you might find using those search terms:
Do you want to start over and get a different recommendation?
Which Statistical Analysis Should You Use?
Recommended for You! Based on what you are researching and the number and types of variables you plan to include, you could use a: Factorial Ancova
Do you want to know more about this type of analysis?
Capella’s Quantitative Skills Center on iGuide provides helpful statistical resources and support, such as Choosing the Correct Statistical Test. This hour-long presentation describes various types of statistical analyses and offers detailed information and guidance.
The Web also has many resources that can help you understand, structure and conduct your statistical analysis. Try searching for them using the name of the recommended statistical analysis and the types of variables you will use as your search terms. Here is an example of the resources you might find using those search terms:
Do you want to start over and get a different recommendation?
Which Statistical Analysis Should You Use?
Recommended for You! Based on what you are researching and the number and types of variables you plan to include, you could use a: One-Way Manova or Hotelling's T-Squared
Do you want to know more about this type of analysis?
Capella’s Quantitative Skills Center on iGuide provides helpful statistical resources and support, such as Choosing the Correct Statistical Test. This hour-long presentation describes various types of statistical analyses and offers detailed information and guidance.
The Web also has many resources that can help you understand, structure and conduct your statistical analysis. Try searching for them using the name of the recommended statistical analysis and the types of variables you will use as your search terms. Here is an example of the resources you might find using those search terms:
- Multivariate Analysis of Variance (MANOVA)
- Annotated SPSS Output: One-Way Manova
- Hotelling's T-Squared Test
Do you want to start over and get a different recommendation?
Which Statistical Analysis Should You Use?
Recommended for You! Based on what you are researching and the number and types of variables you plan to include, you could use a: One-Way Mancova
Do you want to know more about this type of analysis?
Capella’s Quantitative Skills Center on iGuide provides helpful statistical resources and support, such as Choosing the Correct Statistical Test. This hour-long presentation describes various types of statistical analyses and offers detailed information and guidance.
The Web also has many resources that can help you understand, structure and conduct your statistical analysis. Try searching for them using the name of the recommended statistical analysis and the types of variables you will use as your search terms. Here is an example of the resources you might find using those search terms:
Do you want to start over and get a different recommendation?
Which Statistical Analysis Should You Use?
Recommended for You! Based on what you are researching and the number and types of variables you plan to include, you could use a: Factorial Manova
Do you want to know more about this type of analysis?
Capella’s Quantitative Skills Center on iGuide provides helpful statistical resources and support, such as Choosing the Correct Statistical Test. This hour-long presentation describes various types of statistical analyses and offers detailed information and guidance.
The Web also has many resources that can help you understand, structure and conduct your statistical analysis. Try searching for them using the name of the recommended statistical analysis and the types of variables you will use as your search terms. Here is an example of the resources you might find using those search terms:
Do you want to start over and get a different recommendation?
Which Statistical Analysis Should You Use?
Recommended for You! Based on what you are researching and the number and types of variables you plan to include, you could use a: Factorial Mancova
Do you want to know more about this type of analysis?
Capella’s Quantitative Skills Center on iGuide provides helpful statistical resources and support, such as Choosing the Correct Statistical Test. This hour-long presentation describes various types of statistical analyses and offers detailed information and guidance.
The Web also has many resources that can help you understand, structure and conduct your statistical analysis. Try searching for them using the name of the recommended statistical analysis and the types of variables you will use as your search terms. Here is an example of the resources you might find using those search terms:
Do you want to start over and get a different recommendation?
Are you struggling with this?
If something doesn't make sense or you need a little more help, please reach out to our Statistical Help Desk by calling 123-456-7890 or emailing statshelp@capella.edu.
Which Statistical Analysis Should You Use?
Recommended for You! Based on what you are researching and the number and types of variables you plan to include, you could use a: Pearson's R
Do you want to know more about this type of analysis?
Capella’s Quantitative Skills Center on iGuide provides helpful statistical resources and support, such as Choosing the Correct Statistical Test. This hour-long presentation describes various types of statistical analyses and offers detailed information and guidance.
The Web also has many resources that can help you understand, structure and conduct your statistical analysis. Try searching for them using the name of the recommended statistical analysis and the types of variables you will use as your search terms. Here is an example of the resources you might find using those search terms:
Do you want to start over and get a different recommendation?
Which Statistical Analysis Should You Use?
Recommended for You! Based on what you are researching and the number and types of variables you plan to include, you could use a: Spearman's Rho
Do you want to know more about this type of analysis?
Capella’s Quantitative Skills Center on iGuide provides helpful statistical resources and support, such as Choosing the Correct Statistical Test. This hour-long presentation describes various types of statistical analyses and offers detailed information and guidance.
The Web also has many resources that can help you understand, structure and conduct your statistical analysis. Try searching for them using the name of the recommended statistical analysis and the types of variables you will use as your search terms. Here is an example of the resources you might find using those search terms:
Do you want to start over and get a different recommendation?
Which Statistical Analysis Should You Use?
Recommended for You! Based on what you are researching and the number and types of variables you plan to include, you could use a: Kendall's Tau
Do you want to know more about this type of analysis?
Capella’s Quantitative Skills Center on iGuide provides helpful statistical resources and support, such as Choosing the Correct Statistical Test. This hour-long presentation describes various types of statistical analyses and offers detailed information and guidance.
The Web also has many resources that can help you understand, structure and conduct your statistical analysis. Try searching for them using the name of the recommended statistical analysis and the types of variables you will use as your search terms. Here is an example of the resources you might find using those search terms:
Do you want to start over and get a different recommendation?
Which Statistical Analysis Should You Use?
Recommended for You! Based on what you are researching and the number and types of variables you plan to include, you could use a: Point Biserial
Do you want to know more about this type of analysis?
Capella’s Quantitative Skills Center on iGuide provides helpful statistical resources and support, such as Choosing the Correct Statistical Test. This hour-long presentation describes various types of statistical analyses and offers detailed information and guidance.
The Web also has many resources that can help you understand, structure and conduct your statistical analysis. Try searching for them using the name of the recommended statistical analysis and the types of variables you will use as your search terms. Here is an example of the resources you might find using those search terms:
Do you want to start over and get a different recommendation?
Which Statistical Analysis Should You Use?
Recommended for You! Based on what you are researching and the number and types of variables you plan to include, you could use a: Phi
Do you want to know more about this type of analysis?
Capella’s Quantitative Skills Center on iGuide provides helpful statistical resources and support, such as Choosing the Correct Statistical Test. This hour-long presentation describes various types of statistical analyses and offers detailed information and guidance.
The Web also has many resources that can help you understand, structure and conduct your statistical analysis. Try searching for them using the name of the recommended statistical analysis and the types of variables you will use as your search terms. Here is an example of the resources you might find using those search terms:
Do you want to start over and get a different recommendation?
Which Statistical Analysis Should You Use?
Recommended for You! Based on what you are researching and the number and types of variables you plan to include, you could use a: Contingency Coefficient
Do you want to know more about this type of analysis?
Capella’s Quantitative Skills Center on iGuide provides helpful statistical resources and support, such as Choosing the Correct Statistical Test. This hour-long presentation describes various types of statistical analyses and offers detailed information and guidance.
The Web also has many resources that can help you understand, structure and conduct your statistical analysis. Try searching for them using the name of the recommended statistical analysis and the types of variables you will use as your search terms. Here is an example of the resources you might find using those search terms:
- Nominal Measures of Correlation: Phi, The Contingency Coefficient, and Cramer's V
- Association Between Variables
Do you want to start over and get a different recommendation?
Which Statistical Analysis Should You Use?
Recommended for You! Based on what you are researching and the number and types of variables you plan to include, you could use a: Cramer's V
Do you want to know more about this type of analysis?
Capella’s Quantitative Skills Center on iGuide provides helpful statistical resources and support, such as Choosing the Correct Statistical Test. This hour-long presentation describes various types of statistical analyses and offers detailed information and guidance.
The Web also has many resources that can help you understand, structure and conduct your statistical analysis. Try searching for them using the name of the recommended statistical analysis and the types of variables you will use as your search terms. Here is an example of the resources you might find using those search terms:
Do you want to start over and get a different recommendation?
Are you struggling with this?
If something doesn't make sense or you need a little more help, please reach out to our Statistical Help Desk by calling 123-456-7890 or emailing statshelp@capella.edu.
Which Statistical Analysis Should You Use?
Recommended for You! Based on what you are researching and the number and types of variables you plan to include, you could use a: Simple Regression
Do you want to know more about this type of analysis?
Capella’s Quantitative Skills Center on iGuide provides helpful statistical resources and support, such as Choosing the Correct Statistical Test. This hour-long presentation describes various types of statistical analyses and offers detailed information and guidance.
The Web also has many resources that can help you understand, structure and conduct your statistical analysis. Try searching for them using the name of the recommended statistical analysis and the types of variables you will use as your search terms. Here is an example of the resources you might find using those search terms:
Do you want to start over and get a different recommendation?
Which Statistical Analysis Should You Use?
Recommended for You! Based on what you are researching and the number and types of variables you plan to include, you could use a: Simple Logistic Regression
Do you want to know more about this type of analysis?
Capella’s Quantitative Skills Center on iGuide provides helpful statistical resources and support, such as Choosing the Correct Statistical Test. This hour-long presentation describes various types of statistical analyses and offers detailed information and guidance.
The Web also has many resources that can help you understand, structure and conduct your statistical analysis. Try searching for them using the name of the recommended statistical analysis and the types of variables you will use as your search terms. Here is an example of the resources you might find using those search terms:
Do you want to start over and get a different recommendation?
Which Statistical Analysis Should You Use?
Recommended for You! Based on what you are researching and the number and types of variables you plan to include, you could use a: Multiple Regression
Do you want to know more about this type of analysis?
Capella’s Quantitative Skills Center on iGuide provides helpful statistical resources and support, such as Choosing the Correct Statistical Test. This hour-long presentation describes various types of statistical analyses and offers detailed information and guidance.
The Web also has many resources that can help you understand, structure and conduct your statistical analysis. Try searching for them using the name of the recommended statistical analysis and the types of variables you will use as your search terms. Here is an example of the resources you might find using those search terms:
Do you want to start over and get a different recommendation?
Which Statistical Analysis Should You Use?
Recommended for You! Based on what you are researching and the number and types of variables you plan to include, you could use a: Logistic Regression
Do you want to know more about this type of analysis?
Capella’s Quantitative Skills Center on iGuide provides helpful statistical resources and support, such as Choosing the Correct Statistical Test. This hour-long presentation describes various types of statistical analyses and offers detailed information and guidance.
The Web also has many resources that can help you understand, structure and conduct your statistical analysis. Try searching for them using the name of the recommended statistical analysis and the types of variables you will use as your search terms. Here is an example of the resources you might find using those search terms:
Do you want to start over and get a different recommendation?
Which Statistical Analysis Should You Use?
Recommended for You! Based on what you are researching and the number and types of variables you plan to include, you could use a: Multinomial Logistic Regression
Do you want to know more about this type of analysis?
Capella’s Quantitative Skills Center on iGuide provides helpful statistical resources and support, such as Choosing the Correct Statistical Test. This hour-long presentation describes various types of statistical analyses and offers detailed information and guidance.
The Web also has many resources that can help you understand, structure and conduct your statistical analysis. Try searching for them using the name of the recommended statistical analysis and the types of variables you will use as your search terms. Here is an example of the resources you might find using those search terms:
Do you want to start over and get a different recommendation?
Are you struggling with this?
If something doesn't make sense or you need a little more help, please reach out to our Statistical Help Desk by calling 123-456-7890 or emailing statshelp@capella.edu.