{"id":75161,"date":"2021-07-07T15:53:09","date_gmt":"2021-07-07T13:53:09","guid":{"rendered":"https:\/\/www.scribbr.nl\/?p=75161"},"modified":"2023-06-22T10:24:50","modified_gmt":"2023-06-22T08:24:50","slug":"correlational-research","status":"publish","type":"post","link":"https:\/\/www.scribbr.com\/methodology\/correlational-research\/","title":{"rendered":"Correlational Research | When & How to Use"},"content":{"rendered":"

A correlational research design<\/a> investigates relationships between variables<\/a> without the researcher controlling or manipulating any of them.<\/p>\n

A correlation reflects the strength and\/or direction of the relationship between two (or more) variables. The direction of a correlation can be either positive or negative.<\/p>\n\n\n\n\n\n
Positive correlation<\/th>\nBoth variables change in the same direction<\/td>\nAs height<\/span> increases, weight<\/span> also increases<\/td>\n<\/tr>\n
Negative correlation<\/th>\nThe variables change in opposite directions<\/td>\nAs coffee consumption<\/span> increases, tiredness<\/span> decreases<\/td>\n<\/tr>\n
Zero correlation<\/th>\nThere is no relationship between the variables<\/td>\nCoffee consumption<\/span> is not correlated with height<\/span><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

<\/p>\n

Correlational vs. experimental research<\/h2>\n

Correlational and experimental research<\/a> both use quantitative methods<\/a> to investigate relationships between variables. But there are important differences in data collection<\/a> methods and the types of conclusions you can draw.<\/p>\n\n\n\n\n\n\n\n\n
<\/td>\nCorrelational research<\/th>\nExperimental research<\/th>\n<\/tr>\n<\/thead>\n
Purpose<\/th>\nUsed to test strength of association between variables<\/td>\nUsed to test cause-and-effect relationships between variables<\/td>\n<\/tr>\n
Variables<\/th>\nVariables are only observed with no manipulation or intervention by researchers<\/td>\nAn independent variable<\/a> is manipulated and a dependent variable is observed<\/td>\n<\/tr>\n
Control<\/th>\nLimited control<\/a> is used, so other variables may play a role in the relationship<\/td>\nExtraneous variables<\/a> are controlled so that they can\u2019t impact your variables of interest<\/td>\n<\/tr>\n
Validity<\/th>\nHigh external validity<\/a>: you can confidently generalize your conclusions to other populations or settings<\/td>\nHigh internal validity<\/a>: you can confidently draw conclusions about causation<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

When to use correlational research<\/h2>\n

Correlational research is ideal for gathering data quickly from natural settings. That helps you generalize<\/a> your findings to real-life situations in an externally valid way.<\/p>\n

There are a few situations where correlational research is an appropriate choice.<\/p>\n

To investigate non-causal relationships<\/h3>\n

You want to find out if there is an association between two variables, but you don\u2019t expect to find a causal relationship between them.<\/p>\n

Correlational research can provide insights into complex real-world relationships, helping researchers develop theories and make predictions.<\/p>\n

Example<\/figcaption>You want to know if there is any correlation between the number of children people have and which political party they vote for. You don\u2019t think having more children causes people to vote differently\u2014it\u2019s<\/a> more likely that both are influenced by other variables such as age, religion, ideology and socioeconomic status. But a strong correlation could be useful for making predictions about voting patterns.<\/figure>\n

To explore causal relationships between variables<\/h3>\n

You think there is a causal relationship between two variables, but it is impractical, unethical, or too costly to conduct experimental research that manipulates one of the variables.<\/p>\n

Correlational research can provide initial indications or additional support for theories about causal relationships.<\/p>\n

Example<\/figcaption>You want to investigate whether greenhouse gas emissions cause global warming. It is not practically possible to do an experiment that controls global emissions over time, but through observation and analysis you can show a strong correlation that supports the theory.<\/figure>\n

To test new measurement tools<\/h3>\n

You have developed a new instrument for measuring your variable, and you need to test its reliability or validity<\/a>.<\/p>\n

Correlational research can be used to assess whether a tool consistently or accurately captures the concept it aims to measure.<\/p>\n

Example<\/figcaption>You develop a new scale to measure loneliness in young children based on anecdotal evidence during lockdowns. To validate this scale, you need to test whether it\u2019s actually measuring loneliness. You collect data on loneliness using three different measures, including the new scale, and test the degrees of correlations between the different measurements. Finding high correlations means that your scale is valid.<\/figure>\n

How to collect correlational data<\/h2>\n

There are many different methods<\/a> you can use in correlational research. In the social and behavioral sciences, the most common data collection methods for this type of research include surveys, observations<\/a>, and secondary data.<\/p>\n

It\u2019s important to carefully choose and plan your methods to ensure the reliability and validity of your results. You should carefully select a representative sample<\/a> so that your data reflects the population you\u2019re interested in without research bias<\/a>.<\/p>\n

Surveys<\/h3>\n

In survey research<\/a>, you can use questionnaires<\/a> to measure your variables of interest. You can conduct surveys online, by mail, by phone, or in person.<\/p>\n

Surveys are a quick, flexible way to collect standardized data from many participants, but it’s important to ensure that your questions are worded in an unbiased way and capture relevant insights.<\/p>\n

Example<\/figcaption>To find out if there is a relationship between vegetarianism and income, you send out a questionnaire about diet to a sample<\/a> of people from different income brackets. You statistically analyze the responses to determine whether vegetarians generally have higher incomes.<\/figure>\n

Naturalistic observation<\/h3>\n

Naturalistic observation<\/a> is a type of field research where you gather data about a behavior or phenomenon in its natural environment.<\/p>\n

This method often involves recording, counting, describing, and categorizing actions and events. Naturalistic observation can include both qualitative and quantitative<\/a> elements, but to assess correlation, you collect data that can be analyzed quantitatively<\/a> (e.g., frequencies, durations, scales, and amounts).<\/p>\n

Naturalistic observation lets you easily generalize your results to real world contexts, and you can study experiences that aren\u2019t replicable<\/a> in lab settings. But data analysis can be time-consuming and unpredictable, and researcher bias may skew<\/a> the interpretations.<\/p>\n

Example<\/figcaption>To find out if there is a correlation between gender and class participation, you observe college seminars, note the frequency and duration of students\u2019 contributions, and categorize them based on gender. You statistically analyze the data to determine whether men are more likely to speak up in class than women.<\/figure>\n

Secondary data<\/h3>\n

Instead of collecting original data, you can also use data that has already been collected for a different purpose, such as official records, polls, or previous studies.<\/p>\n

Using secondary data is inexpensive and fast, because data collection<\/a> is complete. However, the data may be unreliable, incomplete or not entirely relevant, and you have no control over the reliability<\/a> or validity<\/a> of the data collection procedures.<\/p>\n

Example<\/figcaption>To find out if working hours are related to mental health, you use official national statistics and scientific studies from several different countries to combine data on average working hours and rates of mental illness. You statistically analyze the data to see if countries that work fewer hours have better mental health outcomes.<\/figure>\n

How to analyze correlational data<\/h2>\n

After collecting data, you can statistically analyze the relationship between variables using correlation or regression<\/a> analyses, or both. You can also visualize the relationships between variables with a scatterplot.<\/p>\n

Different types of correlation coefficients and regression analyses are appropriate for your data based on their levels of measurement<\/a> and distributions<\/a>.<\/p>\n

Correlation analysis<\/h3>\n

Using a correlation analysis, you can summarize the relationship between variables into a correlation coefficient<\/strong><\/a>: a single number that describes the strength and direction of the relationship between variables. With this number, you\u2019ll quantify the degree of the relationship between variables.<\/p>\n

The Pearson product-moment correlation coefficient<\/a>, also known as Pearson\u2019s r<\/em>, is commonly used for assessing a linear relationship between two quantitative variables.<\/p>\n

Correlation coefficients are usually found for two variables at a time, but you can use a multiple correlation coefficient for three or more variables.<\/p>\n

Regression analysis<\/h3>\n

With a regression analysis<\/a>, you can predict how much a change in one variable will be associated with a change in the other variable. The result is a regression equation<\/strong> that describes the line on a graph of your variables.<\/p>\n

You can use this equation to predict the value of one variable based on the given value(s) of the other variable(s). It\u2019s best to perform a regression analysis after testing for a correlation between your variables.<\/p>\n

Correlation and causation<\/h2>\n

It\u2019s important to remember that correlation does not imply causation<\/a>. Just because you find a correlation between two things doesn\u2019t mean you can conclude one of them causes the other for a few reasons.<\/p>\n

Directionality problem<\/h3>\n

If two variables are correlated, it could be because one of them is a cause and the other is an effect. But the correlational research design doesn\u2019t allow you to infer<\/a> which is which. To err on the side of caution, researchers don\u2019t conclude causality from correlational studies.<\/p>\n

Example<\/figcaption>You find a positive correlation between vitamin D levels and depression: people with low vitamin D levels are more likely to have depression. But you can\u2019t be certain about whether having low vitamin D levels causes depression, or whether having depression causes reduced intakes of vitamin D through lifestyle or appetite changes. Therefore, you can only conclude that there is a relationship between these two variables.<\/figure>\n

Third variable problem<\/h3>\n

A confounding variable<\/a> is a third variable that influences other variables to make them seem causally related even though they are not. Instead, there are separate causal links between the confounder and each variable.<\/p>\n

In correlational research, there\u2019s limited or no researcher control over extraneous variables<\/a>. Even if you statistically control for some potential confounders, there may still be other hidden variables that disguise the relationship between your study variables.<\/p>\n

Example<\/figcaption>You find a strong positive correlation between working hours and work-related stress: people with lower working hours report lower levels of work-related stress. However, this doesn\u2019t prove that lower working hours causes a reduction in stress.<\/p>\n

There are many other variables that may influence both variables, such as average income, working conditions, and job insecurity. You might statistically control for these variables, but you can\u2019t say for certain that lower working hours reduce stress because other variables may complicate the relationship.<\/figure>\n

Although a correlational study can\u2019t demonstrate causation on its<\/a> own, it can help you develop a causal hypothesis<\/a> that\u2019s tested in controlled experiments.<\/p>\n

Other interesting articles<\/h2>\n

If you want to know more about statistics<\/a>, methodology<\/a>, or research bias<\/a>, make sure to check out some of our other articles with explanations and examples.<\/p>\n

\n
\n
\n

<\/em>Statistics<\/strong><\/p>\n