Analysis on Dating Styles
Paige Saengsouvanna | Data Scientist | UCSB Alum | BS in Data Science and Statistics
Background
Navigating the dating world as young adults can be confusing and frustrating. Some of us are dating for fun, some for marriage, and some are scared of lifelong commitment. I gathered data to explore the various dating styles my peers embody, aiming to provide insight for respondents to better understand their position relative to others in the survey. By the way, this project is exploratory in nature, so I’m not necessarily trying to prove anything—just seeing if there are interesting trends :)
Drawing from my socialization studies, I’ve come to learn that trial and success > trial and error. In other words, you might learn from your mistakes, but you ultimately learn the most when you succeed. In the context of this project, I define success as a meaningful relationship. Defining what constitutes a “meaningful” relationship may vary from person to person. For me, it’s fairly straightforward, like a binary question. If you can wholeheartedly affirm that a relationship brought you no lessons or benefits, then it wasn’t meaningful. Otherwise, if the connection led to personal growth, I’d say that’s considered meaningful. Some people may be lying to themselves about who was meaningful to them, but deep down, THEY KNOW.
The way someone approaches dating can significantly influence their willingness to continue a connection. Those leaning towards the more serious side of dating are often more reluctant to tolerate undesired behavior and are less inclined to consider second chances if they cannot foresee a long-term future. There is also a strong emphasis on compatibility and shared goals.
On the other hand, individuals adopting a more laid-back and casual attitude towards dating tend to prioritize enjoyment and spontaneity. For them, the dating experience is about having a good time and forming connections, often with a willingness to overlook (🚩🫢alarming🫢🚩) personality traits that may not align perfectly with their own. The pursuit of fun and enjoyment takes precedence over long-term compatibility. This approach can result in a more relaxed and flexible approach to the dating experience, allowing for a broader spectrum of interactions and connections.
Due to resource constraints, I shared my survey with my Instagram followers, friends, and classmates, the majority of whom are college students in California. Given the lack of representativeness in my sample, there is inherent bias, and my results cannot be generalized to any specific population. In other words, take these findings with a grain of salt. What I find may or may not translate to your circle of friends/classmates. The purpose of this study is to simply explore the results of the survey. The technical skills I use are formal, but since this is a FUN personal project, my demeanor and tone in this report will be more casual and conversational.
In my survey, I presented a series of questions to my followers, seeking information about their age, the number of total connections they’ve made, and the number of meaningful connections. Respondents self-categorized into seven different dating styles. The self-categorization introduces a degree of bias, but this is for fun. In part of the survey, I asked:
Where are you on the spectrum of non-serious to serious dating style? I understand we can go through multiple phases of dating styles in our lives, but try to choose a number that best describes your dating style history. How do you usually approach dating?
- free spirit
- I typically do not search for any form of commitment
- I date in search of pleasure and fun
- I do not want a relationship
- possible examples: one-night stands, flings, a bae for the night
- the only feelings I want to feel are an orgasm
- AKA “friends with benefits”
- I typically have relationships with people that can involve casual sex, non-exclusivity, fun and pleasure
- not a romantic relationship, more of a sexual relationship
- here for a good time, and if I like you enough, maybe for a long time (not really that long usually)
- AKA “situation-ships”
- I typically have relationships with people that can involve casual sex, connection, fun and pleasure
- can mimic emotional vulnerability of a monogamous relationship, without the monogamy
- go-er with the flow-er
- CAN be a mix of 3 and 5 dating style
- not actively searching for a relationship or situation-ship, hence the name, I go with the flow
- maybe they’re the one! or at least one of the ones!
- I am typically a relationship person, but I date mostly for fun (short or long term relationships)
- romantic and/or sexual relationship
- possible reasons: young, want/able to be monogamous but not thinking about marriage, want to avoid STDs by having one sexual partner at a time (however, the terms of exclusivity is subjective to each relationship)
- this includes serial daters: person who jumps from one relationship to another
- cohabiter
- I date to date and then after a period of time, I will/can consider marriage
- cohabitation: a long-term relationship, can be living together, have sexual and/or romantic relationship, but NOT married
- common amongst young people as well
- 4lifer. i don’t want anyone else, but you!
- dating to marry
- I am only going to date someone if I think I will eventually marry them
- when I date someone, I think about my future with them as my life partner
- a deal breaker if I do not see a future with them
Exploratory Analysis
Below, you’ll find the initial ten rows of the dataset and there are no missing data entries 😁. The data set comprises 107 observations across five variables: timestamp, age, level, total_connections, and meaningful_connections. The summary statistics compute the min, 1st quantile (25th percentile), median (50th percentile), mean, 3rd quantile (75th percentile), and max of each variable. Our summary statistics are unreasonable and not beneficial for analysis. Well, you can’t trust everyone to fill out a survey truthfully nowadays😔, which is a compelling reason enough to remove outliers.
Variable Description
- timestamp: Time respondent answered the survey
- age: The age of respondent
- level: Self-categorized dating style level
- total_connections: Total connections
- meaningful_connections: total meaningful connections
| timestamp | age | level | total_connections | meaningful_connections |
|---|---|---|---|---|
| 2022/12/22 4:34:51 PM PST | 21 | 5 | 2 | 1 |
| 2022/12/22 4:35:49 PM PST | 21 | 5 | 8 | 4 |
| 2022/12/22 4:37:41 PM PST | 80 | 2 | 10000 | 1 |
| 2022/12/22 4:38:05 PM PST | 20 | 6 | 3 | 2 |
| 2022/12/22 4:40:48 PM PST | 21 | 6 | 6 | 1 |
| 2022/12/22 4:42:53 PM PST | 21 | 3 | 7 | 3 |
| 2022/12/22 4:43:18 PM PST | 21 | 6 | 3 | 1 |
| 2022/12/22 4:49:14 PM PST | 30 | 1 | 30 | 30 |
| 2022/12/22 4:55:18 PM PST | 23 | 6 | 10 | 5 |
| 2022/12/22 5:03:05 PM PST | 21 | 7 | 5 | 3 |
| timestamp | age | level | total_connections | meaningful_connections | |
|---|---|---|---|---|---|
| Length:107 | Min. :16.00 | Min. :1.000 | Min. : 0.0 | Min. : 0.000 | |
| Class :character | 1st Qu.:20.00 | 1st Qu.:4.000 | 1st Qu.: 3.0 | 1st Qu.: 2.000 | |
| Mode :character | Median :21.00 | Median :4.000 | Median : 6.0 | Median : 3.000 | |
| NA | Mean :21.57 | Mean :4.542 | Mean : 106.1 | Mean : 4.187 | |
| NA | 3rd Qu.:22.00 | 3rd Qu.:6.000 | 3rd Qu.: 11.5 | 3rd Qu.: 4.000 | |
| NA | Max. :80.00 | Max. :7.000 | Max. :10000.0 | Max. :30.000 |
Below is a subset of the data that is considered to be outliers. While some outliers may genuinely reflect a respondent’s experience, they are outlandish relative to the rest of our data and can significantly skew our findings. In this case, there were 18 such outliers, which I’ve removed. You can see that some of these responses are from trolls. This process leaves us with a total of 92 observations for our analysis.
| timestamp | age | level | total_connections | meaningful_connections |
|---|---|---|---|---|
| 2022/12/22 4:37:41 PM PST | 80 | 2 | 10000 | 1 |
| 2022/12/22 4:49:14 PM PST | 30 | 1 | 30 | 30 |
| 2022/12/22 5:43:01 PM PST | 26 | 5 | 3 | 1 |
| 2022/12/23 8:17:32 PM PST | 30 | 1 | 30 | 30 |
| 2022/12/24 7:57:01 PM PST | 16 | 4 | 10 | 3 |
| 2022/12/22 4:37:41 PM PST | 80 | 2 | 10000 | 1 |
| 2022/12/22 4:49:14 PM PST | 30 | 1 | 30 | 30 |
| 2022/12/22 11:38:51 PM PST | 23 | 5 | 60 | 4 |
| 2022/12/23 1:58:24 AM PST | 21 | 4 | 50 | 15 |
| 2022/12/23 5:53:31 PM PST | 23 | 5 | 30 | 3 |
| observation | age | level | total_connections | meaningful_connections | |
|---|---|---|---|---|---|
| Min. : 1.00 | Min. :17.00 | Min. :1.000 | Min. : 0.000 | Min. : 0.000 | |
| 1st Qu.:23.75 | 1st Qu.:20.00 | 1st Qu.:4.000 | 1st Qu.: 3.000 | 1st Qu.: 2.000 | |
| Median :46.50 | Median :21.00 | Median :5.000 | Median : 5.000 | Median : 2.000 | |
| Mean :46.50 | Mean :20.74 | Mean :4.674 | Mean : 6.946 | Mean : 3.076 | |
| 3rd Qu.:69.25 | 3rd Qu.:22.00 | 3rd Qu.:6.000 | 3rd Qu.: 9.250 | 3rd Qu.: 4.000 | |
| Max. :92.00 | Max. :24.00 | Max. :7.000 | Max. :20.000 | Max. :15.000 |
After eliminating the outlandish responses, the age range is now from 17 to 24 years old. The total connections range from 0 to 20, and the meaningful connections variable spans from 0 to 15. These revised ranges are more aligned with my expectations; I now have a more reasonable dataset compared to before 😁😁😁😁😁😁😁
Age and Meaningful Connections
The scatter plot between age and meaningful connections shows a very weak positive correlation. As age increases, we observe a corresponding increase in the number of meaningful connections, but, again, this correlation is weak. The scatter plot between age and total connections reveals that there isn’t a clear trend in the number of total connections formed as individuals age. So, age doesn’t seem to have a significant impact on the number of total and meaningful connections.
Age and Dating Style
The frequency distribution of the age variable is organized based on dating style levels. There is a higher representation of 21-year-olds in the dataset compared to other age groups, which isn’t surprising since most of my friends who participated in the survey are 21. A more substantial sample across various age groups would reveal more distinct patterns, but we’ll make do with what we have. It’s interesting to see that 18-year-olds are leaning towards the more serious end of the dating spectrum, signified by the prominent pink block indicating their interest in dating for marriage 😳😳. I want to say I’m surprised, but it’s almost a rite of passage. Were you really 18 years old if you didn’t think you were going to marry your high school boyfriend/girlfriend? 😜 Reality will hit them sooner or later. Interestingly, a noticeable, yet small percentage of individuals aged 18 to 22 express an intention to marry through their dating style.
In the second chart, I’ve inverted the variables, allowing us to examine the frequency distribution of dating style levels grouped by age. Six individuals identify as level 1, while NOT A SINGLE ONE fell into level 2 — 😂😂😂. Looking back, it’s possible that the title (“the only feelings I want to feel are an orgasm”) might have intimidated potential respondents (lol). Reflecting on this, it’s my personal belief that our generation has evolved from the “friends with benefits” relationship to a more nuanced dynamic, AKA “situationships”. We can see that there are more individuals on the serious side of the dating spectrum (5-7) than on the non-serious side (1-3), indicating a significant number of individuals pursuing deeper emotional connections in this dataset. We have some lover girls/boys in this dataset 💝🤭.
Total Connections and Meaningful Connections
Here is the frequency distribution of total connections categorized by dating style. Individuals seeking a long-term commitment (Level 7) tend to have fewer total connections. This makes sense, as those with marriage in mind tend to be more selective about their dating choices.
Those falling into the 16-20 total connections group embody less serious dating styles. Level 5 represents a significant proportion, and this may seem contradictory since Level 5 is on the more serious end of the spectrum. However, we must remember that Level 5 includes the 🔪🔪serial daters🔪🔪. These individuals love being in a relationship, enjoy the romance, and value companionship. They are considered serious in their dating style because they tend to invest a substantial amount of time in their relationships. Nevertheless, when their relationship doesn’t work out, they’re in search of a new lover, and in the process, they’re dating… and dating… and dating people to find their next “the one”. We all know someone like this 😜😜.
Level 4 represents those who adopt a “go-with-the-flow” approach to dating. They date casually and sporadically, without actively seeking a specific outcome. It’s not surprising to see Level 4 distributed across various intervals for total connections, as these individuals do not hold rigid expectations. Since they occupy a middle ground between non-serious and serious dating styles, it makes sense that a considerable number of respondents fall into this category.
Success Rate
The scatter plot examining the relationship between level and success rate doesn’t reveal any apparent correlation or distinctive patterns. In the subsequent graph, we shift our focus to explore the distribution of success rates across different dating style levels. This visualization shows how success rates are distributed within each level. By categorizing the data in this fashion, we gain insights into the dating experiences of our survey participants, and I hope it’s easier to interpret :) A significant proportion of respondents in levels 5-7 achieved a relatively high success rate (0.51-1.00). Specifically, a majority of respondents who are pursuing marriage (pink/7) are rather successful. The proportions of success (0.51-1.00) and failures (0.00-0.50) in levels 4 and 5 are close to even.
Correlation Matrix
The correlation matrix measures correlation (ranging from -1 to 1) and linearity between pairs of variables. In this analysis, the categorical variable “dating level” was excluded from consideration. The correlation coefficient between “meaningful connections” and “total connections” stands at 0.6, indicating a moderate positive correlation. This suggests that as the number of total connections increases, there is a corresponding increase in meaningful connections, and vice versa. This relationship is intuitive; it’s reasonable to assume that more social interactions would naturally result in a greater likelihood of forming meaningful connections.
Conversely, the correlation coefficient between “meaningful connections” and “success rate” is 0.45. This moderate positive correlation reflects the definition of success rate as the proportion of meaningful connections to total connections. Consequently, the more meaningful connections an individual has, the higher their success rate. Hence, success rate and total connections have a negative value. Other correlations are weak or close to negligible.
Success Rate and Dating Style
To delve deeper into the relationship between success rate and dating style, a binary categorization was employed. A success rate ranging from 0 to 0.50 was classified as “No Success,” while a success rate between 0.51 and 1.0 was labeled as “Success.” I hope you find this easier to interpret 😁. With this new categorization, there are more “No Success” instances across all dating styles. In other words, there were more “fails” than successes when using this criterion.
| 1 | 3 | 4 | 5 | 6 | 7 | |
|---|---|---|---|---|---|---|
| No Success | 5 | 9 | 14 | 13 | 10 | 8 |
| Success | 1 | 4 | 11 | 5 | 5 | 7 |
Lastly, I am going to utilize an ANOVA test, which assesses the means of the success rate across the 7 dating style levels (I’ve already checked for its assumptions). The question is: We can see some stark differences in the contingency table between the dating style levels, but is there enough data to suggest these differences are significant?
## Df Sum Sq Mean Sq F value Pr(>F)
## level 1 0.098 0.09815 1.514 0.222
## Residuals 90 5.835 0.06484H0: There are no significant differences in the means of dating styles being compared (group means are equal) Ha: There is a significant difference in atleast one of the mean success rates of dating styles being compared is not equal to the others
We get a p-value of 0.222, so we fail to reject null, and there is not enough evidence to conclude there is a significant difference in the means of dating styles being compared. (**Note: If we had more respondents in general across the different dating styles, the test could have came out differently)
THE END THANKS 💝
I hope this relieved some pressure, knowing that it may not really matter to abide by certain dating behaviors. I found it really sweet to discover that most respondents are dating in pursuit of an emotional connection 🥂💕.
A part of being a data scientist is tailoring your analysis to your audience. There are many ways to analyze data, using different statistical tests, modeling, etc. However, since my audience is my Instagram followers, I assume that they are only interested in graphs. Since this is for fun, I am opting out of doing any further analysis. Typically, analysis is very objective and unbiased, but that can be boring, and it’s a lot more fun to weigh in my personal opinions and experiences. I hope you enjoyed seeing how the results turned out. Thank you to everyone who participated in this study 😘😘.
P.S I am sorry for taking so long to put this out. I have been severely burnt out.
Paige 🖤🖤