A predictor variable is used to predict the occurrence and/or level of another variable, called the outcome variable.
A researcher will measure both variables in a scientific study and then use statistical software to determine if the predictor variable is associated with the outcome variable. If there is a strong correlation, we say the predictor variable has high predictive validity.
This methodology is often used in epidemiological research. Researchers will measure both variables in a given population and then determine the degree of association between the predictor and outcome variable.
This allows scientists to examine the connection between many meaningful variables, such as exercise and health or personality type and depression, just to give a few examples.
Although this type of research can provide significant insights that help us understand a phenomenon, we cannot say that the predictor valuable causes the outcome variable.
In order to use the term ‘cause and effect’, the researcher must be able to control and manipulate the level of a variable and then observe the changes in the other variable.
Definition of Predictor and Outcome Variables
In reality, many variables usually affect the outcome variable. So, researchers will measure numerous predictor variables in the population under study and then determine the degree of association that each one has with the outcome variable.
It sounds a bit complicated, but fortunately, the use of a statistical technique called multiple regression analysis simplifies the process.
As long as the variables are measured accurately and the population size is large, the software will be able to determine which of the predictor variables are associated with the outcome variable and the degree of association.
Not all predictors will have an equal influence on the outcome variable. Some may have a very small impact, some may have a substantial impact, and others may have no impact at all.
Predictor and outcome are not to be confused with independent and dependent variables.
Examples of Predictor and Outcome Variables
1. Diet and Health
Does the food you eat have any impact on your physical health? This is a question that a lot of people want to know the answer to.
Many of us have very poor diets, with lots of fast food and salty snacks. Other people, however, almost never make a run through the drive-thru, and consume mostly fruits and veggies.
Thankfully, epidemiological research can give us a relatively straightforward answer. First, researchers measure the quality of diet of each person in a large population.
So, they will track how much fast food and fruits and veggies people consume. There are a lot of different ways to measure this.
Secondly, researchers will measure some aspects of health. This could involve checking cholesterol levels, for example. There are a lot of different ways to measure health. The final step is to input all of the data into the statistical software program and perform the regression analysis to see the results.
Quality of diet is the predictor variable, and health is the outcome variable.
2. Noise Pollution and IQ
One scientist speculates that living in a noisy environment will affect a person’s ability to concentrate, which will then affect their mental acuity and subsequent cognitive development.
So, they decide to conduct a study examining the relationship between noise pollution and IQ.
First, they travel through lots of different neighborhoods and use a sound level meter to assess noise pollution. Some neighborhoods are in the suburbs, and some are near busy highways or construction sites.
Next, they collect data on SAT scores of the children living in those neighborhoods.
They then conduct a regression analysis to determine the connection between the sound level meter data and the SAT scores.
In this example, the predictor variable is the sound levels, and the outcome variable is the SAT scores.
Surprisingly, the results revealed an inverse relationship between noise and SAT scores. That is, the more noise in the environment the higher the SAT score. Any idea why?
3. Family Income and Achievement Test Scores
In this study, sociologists conducted a study examining the relationship between how much income a family has and the achievement test scores of their children.
The researchers collected data from schools on the achievement test scores of hundreds of students and then estimated the household income of the families based on the occupation of the parents.
The results revealed a strong relationship between family income and test scores, such that the higher the family income, the higher the test score of the child.
In this example, family income is the predictor variable, and test score is the outcome variable.
4. Parental Utterances and Children’s Vocabulary
A team of child psychologist is interested in the impact of how much parents talk to their child and that child’s verbal skills.
So, they design a study that involves observing families in the home environment. They randomly choose 50 families to study that live nearby.
A research assistant visits each family, records, and later counts the number of utterances spoken by the mother directed at their only child.
On a different occasion, a second research assistant administers a verbal skills test to every child. Yes, this type of study takes a lot of time.
The regression analysis reveals a direct relationship between the number of utterances from the mother and the child’s verbal skills test score. The more utterances, the higher the score.
In this example, the predictor variable is the number of utterances directed at the child, and the outcome variable is the child’s verbal skills test score.
5. Video Games and Aggressiveness
The debate about the effects of TV violence and video games has been raging for nearly 70 years. There have been hundreds, maybe even thousands of studies conducted on the issue.
One type of study involves assessing how frequently a group of people play certain video games and then tracking their level of aggressiveness over a period of time.
Of course, there are other factors involved in whether a person is aggressive or not, so the researchers might assess those variables as well.
In this type of study, the predictor variable is the frequency of playing video games, and the outcome variable is the level of aggressiveness.
6. Chemicals in Food Products and Puberty
In many countries, farmers may inject various antibiotics and growth hormones into their cattle to ward off infection and increase body mass and milk production.
Unfortunately, those chemicals do not disappear once the food hits the supermarket shelves. Some parents, educators, and food scientists began to notice an association between these agricultural practices and the onset of puberty in young children.
Numerous scientific studies were conducted examining the relationship between these practices and puberty.
So, the researchers studied the relationship between the predictor variable (chemicals in food) and the outcome variable (onset of puberty).
7. Full Moon and Craziness
Who hasn’t heard that a full moon brings out the crazies? A lot of people have theorized that when the moon is full, people get a little bit wild and uninhibited.
That can lead to people doing things they would not normally do.
To put this theory to the test, a group of criminologists decides to examine the police records of numerous large cities and compare that with the lunar cycle.
The researchers input all of the data into a stats program to examine the degree of association between police incidents and the moon.
In this study, the lunar cycle is the predictor variable, and contravention of the law is the outcome variable.
8. Testosterone and Leadership Style
There are many types of leadership styles. Some leaders are very people-oriented and try to help their employees prosper and feel good about their jobs.
Other leaders are more task-driven and prefer to clearly define objectives, set deadlines, and push their staff to work hard.
To examine the relationship between leadership style and testosterone, a researcher first administers a questionnaire to hundreds of employees in several types of companies. The questionnaire asks the employees to describe the leadership style of their primary supervisor.
At the same time, the researcher also collects data on the testosterone levels of those supervisors and matches them with the questionnaire data.
By examining the association between the two, it will be possible to determine if there is a link between leadership style and testosterone.
The predictor variable is testosterone, and the outcome variable is leadership style.
9. Personality Type and Driver Safety
A national bus company wants to hire the safest drivers possible. Fewer accidents mean passengers will be safe and their insurance rates will be lower.
So, the HR staff begin collecting data on the safety records of their drivers over the last 3 years. At the same time, they administer a personality inventory that assesses Type A and Type B personalities.
The Type A personality is intense, impatient, and highly competitive. The Type B personality is easygoing and relaxed. People have varying levels of each type.
The HR department wants to know if there is a relationship between personality type (A or B) and accidents among their drivers.
The predictor variable is personality type, and the outcome variable is the number of accidents.
10. Vitamins and Health
Americans take a lot of vitamins. However, there is some debate about whether vitamins actually do anything to improve health.
There are so many factors that affect health, will taking a daily supplement really count?
So, a group of small vitamin companies pull their resources and hire an outside consulting firm to conduct a large-scale scientific study.
The firm randomly selects thousands of people from throughout the country to participate in the study. The people selected come from a wide range of SES backgrounds, ethnicities, and ages.
Each person is asked to go to a nearby hospital and have a basic health screening that includes cholesterol and blood pressure. They also respond to a questionnaire that asks if they take a multi-vitamin, how many and how often.
The consulting firm then compares the degree of association between multi-vitamins and health.
Multi-vitamin use is the predictor variable, and health is the outcome variable.
11. Automobiles and Climate Change
A group of climatologists has received funding from the EU to conduct a large-scale study on climate change.
The researchers collect data on a wide range of variables that are suspected of affecting the climate. Some of those variables include automobile production, industrial output, size of cattle herds, and deforestation, just to name a few.
The researchers proceed by gathering the data beginning with the 1970s all the way to the current year. They also collect data on yearly temperature fluctuations.
Once all the data is collected, it is put into a stats program, and a few minutes later, the results are revealed.
In this example, there are many predictor variables, such as automobile production, and one primary outcome variable (yearly temperature fluctuations).
12. Smartphone Use and Eye Strain
If you’ve ever noticed, people spend a lot of time looking at their smartphones.
When they are reading, when they are waiting in line, in bed at night, and even when walking from point A to point B.
Many optometrists are concerned that all of this screen time is doing harm to people’s eyesight. So, they decide to conduct a study.
Fortunately, they all work for a nationwide optometry company with offices located in Wal-Marts.
When patients come into their office, they give each one a standard eye exam. They also put a question on the in-take form asking each person to estimate how many hours a day they spend looking at their smartphone screen.
Then they examine the relation between screen-time usage and the results of the eye exams.
In this study, the predictor variable is screen-time, and the outcome variable is the eye-exam results.
13. Soil Composition and Agricultural Yields
Although farming looks easy, it can be a very scientific enterprise. Agriculturalists study the composition of soil to help determine what type of food will grow best.
Today, they know a lot about which soil nutrients affect the growth of different plant varieties because there have been decades of studies.
The research involves collecting soil samples, measuring crop yields, and then examining the association between the two.
For example, scientists will measure the pH levels, mineral composition, as well as water and air content over many acres of land and relate that to the amount harvested of a particular crop (e.g., corn).
In this example, there are numerous predictor variables, all of which have some effect on crop growth, which is the outcome variable.
Even though there are so many variables to consider, the regression analysis will be able to tell us how important each one is in predicting the outcome variable.
There can be a lot of reasons why something happens. More often than not, nothing happens as a result of just one factor. Our physical health, climate change, and a person’s level of aggressiveness are all the result of numerous factors.
Fortunately for science, there is a brilliant way of determining which factors are connected to a phenomenon and how strong is each and every one of them.
By collecting data on a predictor variable (or variables) and then examining the association with the outcome variable, we can gain valuable insights into just about any subject matter we wish to study.
Ferguson, C. J., & Kilburn, J. (2010). Much ado about nothing: The misestimation and overinterpretation of violent video game effects in Eastern and Western nations: Comment on Anderson et al. (2010). Psychological Bulletin, 136(2), 174–178. https://doi.org/10.1037/a0018566
Ferguson, C. J., San Miguel, C., Garza, A., & Jerabeck, J. M. (2012). A longitudinal test of video game violence influences on dating and aggression: A 3-year longitudinal study of adolescents,
Journal of Psychiatric Research, 46(2), 141-146. https://doi.org/10.1016/j.jpsychires.2011.10.014
Gordon, R. (2015). Regression Analysis for the Social Sciences (2nd ed). New York: Routledge.
Hoff, E. (2003). The specificity of environmental influence: Socioeconomic status affects early vocabulary development via maternal speech. Child Development, 74(5), 1368–1378.
Lopez-Rodriguez, D., Franssen, D., Heger, S., & Parent, AS. (2011). Endocrine-disrupting chemicals and their effects on puberty. Best Practice & Research Clinical Endocrinology & Metabolism, 35(5), 101579. https://doi.org/10.1016/j.beem.2021.101579
Man, A., Li, H., & Xia, N. (2020). Impact of lifestyles (Diet and Exercise) on vascular health: Oxidative stress and endothelial function. Oxidative Medicine and Cellular Longevity, 1496462. https://doi.org/10.1155/2020/1496462
Thompson, R., Smith, R. B., Karim, Y. B., Shen, C., Drummond, K., Teng, C., & Toledano, M. B. (2022). Noise pollution and human cognition: An updated systematic review and meta-analysis of recent evidence. Environment International, 158, 106905.
1 thought on “13 Predictor and Outcome Variable Examples”
If I want to undertake an interventional study where I measure the Knowledge, attitudes and practices of adolescents in 3 key sexual and reproductive areas. And their parents’ acceptance of ASRH education for their children, and their misconceptions of ASRH. And then I introduce both children and parents to ASRH education. Then I do an end line to look for improvement in the adolescent’s KAP in those 3 areas, and an increased acceptance of ASRH education among parents, what is my predictor variable and outcome variable?