Introduction
(Slide 1 of 30)

Chapter 15. Tinder and Online Dating: Everything Up to the Pearson Correlation Coefficient

Which Test Is Best?
true
true
You must read each slide, and complete any questions on the slide, in sequence.

By Warren Fass, University of Pittsburgh Bradford, and Susan A. Nolan, Seton Hall University

Basu, T. (2016, March 25). Lessons from Tinder’s in-house love doctor. Opens in new window https://psmag.com/social-justice/odds-are-you-got-on-tinder-to-find-true-love

Smith, A. (2016, February 11). 15% of American adults have used online dating sites or mobile dating apps. Opens in new window https://www.pewinternet.org/2016/02/11/15-percent-of-american-adults-have-used-online-dating-sites-or-mobile-dating-apps/

Asian man is flirting with woman in a bar while having drinks, woman is shy
Kzenon/Shutterstock

In this activity, we will consider various findings from the Pacific Standard article, “Lessons From Tinder’s In-House Love Doctor” (Basu, 2016). Then you will identify statistical analyses that could be used to support the various conclusions.

Guidelines for choosing the appropriate hypothesis test

portrait of natural real african man on blue background
Daxiao Productions/Shutterstock
Attractive young girl
Alberto Zornetta/Shutterstock

Have you ever used an online dating site such as eHarmony, Match.com, or Tinder? According to the results from a survey conducted by the Pew Research Center (Smith, 2016), “15% of American adults have used either online dating sites or mobile dating apps.” One popular dating site, Tinder, is estimated to have over 50 million users. Jessica Carbino was one of those users, while earning her Ph.D. in sociology at UCLA. In fact, one of her Tinder dates was with Sea Rad, the founder of Tinder. The romantic relationship didn’t go anywhere, but as a result of that date, Jessica (now Dr. Carbino) was hired to work at Tinder as an online dating sociologist. In 2016, Dr. Carbino was interviewed by Tanya Basu, who writes for Pacific Standard magazine (Basu, 2016), to answer questions related to her research on facial attractiveness in online dating. What were some of her findings?

Closeup portrait of a confident middle eastern businessman
Jacob Lund/Shutterstock

When looking at profile pictures on Tinder, can facial features tell us about personality characteristics? According to Dr. Carbino, “people are actually able to analyze a photo or snapshot of somebody, whether in person or not in person, and gauge a great deal of information about the person with a lot of accuracy, very fast.” In other words, the judgments occur based upon a small amount of information. Dr. Carbino stated that “women may associate a harder jawline with more brash, more narcissistic tendencies.” Dr. Carbino did not outline the evidence for this theory, but we can design a study to see if there is a relation between jawline and narcissistic tendencies. Suppose we recruit 20 women to participate in the study. The women would be shown a man’s profile picture. The women would be asked to rate the picture according to two different scales. The women would be asked to rate the degree of narcissism (1 = little, if any, 10 = a lot) depicted in that profile picture. The women would also be asked to rate the quality of the jawline of the profile picture (1 = soft, 10 = hard). We would then have two ratings from each woman.

Which statistical test could be used to determine if there was a relation between the two variables – degree of narcissism and quality of jawline?
Correct! The researchers could have used the Pearson correlation coefficient because there are only scale variables, degree of narcissism and quality of jawline. There are no nominal variables. The first scale variable is degree of narcissism on a scale of 1-10, and the second scale variable is quality of jawline on a scale of 1-10.
 
Now you can skip ahead to the next example by clicking here. Or, if you would like more practice walking through the flowchart questions, then simply click the Next button in the bottom right corner of the screen.
Actually, that’s not the correct statistical analysis. Let’s walk through the questions on the flowchart in Appendix E to determine what analysis could be used in this case.

In which of the following four categories does this situation fall? Click to see the data again. And click on the flowchart button to see the overview for choosing the best test.

Suppose we recruit 20 women to participate in the study. The women would be shown a man’s profile picture. The women would be asked to rate the picture according to two different scales. The women would be asked to rate the degree of narcissism (1 = little, if any, 10 = a lot) depicted in that profile picture. The women would also be asked to rate the quality of the jawline of the profile picture (1 = soft, 10 = hard). We would then have two ratings from each woman.
Correct! There are only scale variables. Both narcissism and jawline quality are rated on 1-10 scales.
Actually, there are only scale variables. Both narcissism and jawline quality are rated on 1-10 scales.

Is the research question about association or prediction?
Correct! The research question is about association. We want to know if two variables are related, not whether one predicts the other.
Actually, the research question is about association. We want to know if two variables are related, not whether one predicts the other.

Based on the answers to these questions, which statistical test could be used to determine if there was a relationship between the two variables, degree of narcissism and quality of jawline?
Correct! The researchers could have used the Pearson correlation coefficient because there are only scale variables, degree of narcissism on a scale of 1-10 and quality of jawline on a scale of 1-10, and the research question is about association rather than prediction.
Actually, that’s not the correct statistical analysis. The researchers could have used the Pearson correlation coefficient because there are only scale variables, degree of narcissism on a scale of 1-10 and quality of jawline on a scale of 1-10, and the research question is about association rather than prediction.

Healthy lifestyle concept- happy woman hiker giving thumbs up before summit hike. Beautiful young Asian / Caucasian model. Shows Pico Viejo, Teide National Park, Tenerife with lots of copyspace.
Maridav/Shutterstock

Now let’s look at another finding from Dr. Carbino’s research (Basu, 2016). Dr. Carbino found that a majority of Tinder users wear neutral colors (e.g., white, gray, black) in their photos. Dr. Carbino found that surprising because she expected individuals would wear colors to make them stand out in their photos, perhaps to appear more attractive. Imagine that we recruited 30 adult men, and randomly assigned them to one of three groups—the first group would be shown a picture of an adult woman dressed in a gray outfit, the second group would be shown a picture of an adult woman dressed in a red outfit, and the third group would be shown a picture of an adult woman wearing a green outfit. (The adult woman would be the same person in the three different pictures with the exception of the color of her outfit). After viewing their respective picture, the participants were asked to rate the attractiveness (1 = not very, 10 = very) of the person in the picture.

What statistical analysis could be used to determine whether there was a significant difference among the three groups’ ratings?
Correct! The researchers could have used a one-way between-groups ANOVA because there is one nominal independent variable, color of outfit, with three levels or groups: gray, red, green. There is a scale dependent variable, attractiveness ratings on a scale of 1-10. And participants are only in one of the three groups, so it’s a between-groups design.
 
Now you can skip ahead to the next example by clicking here. Or, if you would like more practice walking through the flowchart questions, then simply click the Next button in the bottom right corner of the screen.
That’s not the correct statistical analysis. Let’s walk through the questions on the flowchart in Appendix E to determine what analysis could be used in this case.

In which of the following four categories does this situation fall? Click to see the data again. And click on the flowchart button to see the overview for choosing the best test.

Imagine that we recruited 30 adult men, and randomly assigned them to one of three groups—the first group would be shown a picture of an adult woman dressed in a gray outfit, the second group would be shown a picture of an adult woman dressed in a red outfit, and the third group would be shown a picture of an adult woman wearing a green outfit. (The adult woman would be the same person in the three different pictures with the exception of the color of her outfit). After viewing their respective picture, the participants were asked to rate the attractiveness (1 = not very, 10 = very) of the person in the picture.
Correct! There is at least one nominal independent variable and a scale dependent variable.
Actually, there is at least one nominal independent variable and a scale dependent variable.

How many nominal independent variables are there?
Correct! There is one nominal independent variable, color of outfit. (The dependent variable, attractiveness ratings, is scale.)
Actually, there is only one nominal independent variable, color of outfit. (The dependent variable, attractiveness ratings, is scale.)

How many levels does the independent variable have?
Correct! There are three levels or groups: gray, red, and green.
Actually, there are three levels or groups: gray, red, and green.

What type of design is this?
Correct! This is a between-groups design. Each participant is assigned to only one of the groups.
Actually, this is a between-groups design. Each participant is assigned to only one of the groups.

Based on the answers to these questions, what statistical analysis could be used to determine whether there was a significant difference among the ratings from the three groups?
Correct! The researchers could have used a one-way between-groups ANOVA because there is one nominal independent variable, color of outfit, with three levels or groups: gray, red, green. There is a scale dependent variable, attractiveness ratings on a scale of 1-10. And participants are only in one of the three groups, so it is a between-groups design.
Actually, the researchers could have used a one-way between-groups ANOVA because there is one nominal independent variable, color of outfit, with three levels or groups: gray, red, green. There is a scale dependent variable, attractiveness ratings on a scale of 1-10. And participants are only in one of the three groups, so it is a between-groups design.

Dr. Carbino stated that one of the benefits of using Tinder is that people must provide a brief bio with their profile picture (Basu, 2016). The brief bio, according to Dr. Carbino, would contain more information about a prospective date, than if you met a person at a bar. Imagine that we recruit a group of 20 men and randomly assign one half of the group to view a profile picture of a woman with her brief bio, and the remaining one half of the group would view the same profile picture without the brief bio. After viewing their respective profile pictures, all of the men were asked to rate the likelihood (1 = not likely, 5 = very likely) they would date the woman in the picture.

What statistical analysis could be used to determine whether there was a significant difference between the likelihood ratings from the two groups?
Correct! The researchers could have used an independent-samples t test because there is one nominal independent variable, presence of bio, with two levels or groups, with bio and without bio. There is a scale dependent variable, likelihood ratings on a scale of 1-5. And participants are only in one of the two groups, so it’s a between-groups design.
 
Now you can skip ahead to the next example by clicking here. Or, if you would like more practice walking through the flowchart questions, then simply click the Next button in the bottom right corner of the screen.
That’s not the correct statistical analysis. Let’s walk through the questions on the flowchart in Appendix E to determine what analysis could be used in this case.

In which of the following four categories does this situation fall? Click to see the data again. And click on the flowchart button to see the overview for choosing the best test.

Imagine that we recruit a group of 20 men and randomly assign one half of the group to view a profile picture of a woman with her brief bio, and the remaining one half of the group would view the same profile picture without the brief bio. After viewing their respective profile pictures, all of the men were asked to rate the likelihood (1 = not likely, 5 = very likely) they would date the woman in the picture.
Correct! There is at least one nominal independent variable and a scale dependent variable.
Actually, there is at least one nominal independent variable and a scale dependent variable.

How many nominal independent variables are there?
Correct! There is one nominal independent variable, presence of bio. (The dependent variable, likelihood rating, is scale.)
Actually, there is only one nominal independent variable, presence of bio. (The dependent variable, likelihood rating, is scale.)

How many levels does the independent variable have?
Correct! The independent variable has two levels or groups, with bio and without bio.
Actually, the independent variable has two levels or groups, with bio and without bio.

How many samples are there?
Correct! There are two samples, one with the bio and one without the bio.
Actually, there are two samples, one with the bio and one without the bio.

What type of design is this?
Correct! This is a between-groups design. Each participant is randomly assigned to only one of the groups.
Actually, this is a between-groups design. Each participant is randomly assigned to only one of the groups.

Based on the answers to these questions, what statistical test could be used to determine whether there was a significant difference between the likelihood ratings from the two groups?
Correct! The researchers could have used an independent-samples t test because there is one nominal independent variable – presence of bio, with two levels or groups, with bio and without bio. There is a scale dependent variable, likelihood ratings on a scale of 1-5. And participants are only in one of the two groups, so it is a between-groups design.
Actually, that’s not the correct statistical analysis. The researchers could have used an independent-samples t test because there is one nominal independent variable – presence of bio, with two levels or groups, with bio and without bio. There is a scale dependent variable, likelihood ratings on a scale of 1-5. And participants are only in one of the two groups, so it is a between-groups design.

Let’s look at one last point from Dr. Carbino’s interview, the belief that Tinder is only used “as a hook-up generator.” Although Dr. Carbino did not compare the use of Tinder to other dating sites, imagine that we conduct the following study. We recruit 20 men, and provide each man only the names of two dating sites: Tinder and Match.com. We ask the men to rate both sites with respect to their agreement (1 = disagree, 5 = agree) about whether that site is used primarily as a hook-up site.

What statistical analysis could be used to determine whether there was a significant difference between the men’s ratings of the two dating sites?
Correct! The researchers could have used a paired-samples t test. There is one nominal independent variable, name of dating site, with two levels or groups, Tinder and Match.com. There is a scale dependent variable, agreement ratings on a scale of 1-5. All participants rated both dating sites, so it is a within-groups design.
 
Now you can skip ahead to the end of the activity by clicking here. Or, if you would like more practice walking through the flowchart questions, then simply click the Next button in the bottom right corner of the screen.
Actually, that’s not the correct statistical analysis. Let’s walk through the questions on the flowchart in Appendix E to determine what analysis could be used in this case.

In which of the following four categories does this situation fall? Click to see the data again. And click on the flowchart button to see the overview for choosing the best test.

Let’s look at one last point from Dr. Carbino’s interview, the belief that Tinder is only used “as a hook-up generator.” Although Dr. Carbino did not compare the use of Tinder to other dating sites, imagine that we conduct the following study. We recruit 20 men, and provide each man only the names of two dating sites: Tinder and Match.com. We ask the men to rate both sites with respect to their agreement (1 = disagree, 5 = agree) about whether that site is used primarily as a hook-up site.
Correct! There is at least one nominal independent variable and a scale dependent variable.
Actually, there is at least one nominal independent variable and a scale dependent variable.

How many nominal independent variables are there?
Correct! There is one nominal independent variable, name of dating site. (The dependent variable, agreement rating, is scale.)
Actually, there is only one nominal independent variable, name of dating site. (The dependent variable, agreement rating, is scale.)

How many levels does the independent variable have?
Correct! The independent variable has two levels or groups, Tinder and Match.com.
Actually, the independent variable has two levels or groups, Tinder and Match.com.

How many samples are there?
Correct! There are two samples, one for the name Tinder and one for the name Match.com.
Actually, there are two samples, one for the name Tinder and one for the name Match.com.

What type of design is this?
Correct! This is a within-groups design. Each participant rated both dating sites.
Actually, this is a within-groups design. Each participant rated both dating sites.

Based on the answers to these questions, what statistical test could be used to determine whether there was a significant difference between the likelihood ratings from the two groups?
Correct! The researchers could have used a paired-samples t test. There is one nominal independent variable, name of dating site, with two levels or groups, Tinder and Match.com. There is a scale dependent variable, agreement ratings on a scale of 1-5. All participants rated both dating sites, so it is a within-groups design.
Actually, that’s not the correct statistical analysis. The researchers could have used a paired-samples t test. There is one nominal independent variable, name of dating site, with two levels or groups, Tinder and Match.com. There is a scale dependent variable, agreement ratings on a scale of 1-5. All participants rated both dating sites, so it is a within-groups design.

Congratulations! You have completed the activity and gained some good experience in choosing the best hypothesis test.