The purpose of this study is to bring knowledge of how couples meet up‐to‐date by asking detailed questions about both the timing and the social contexts of how Americans meet their romantic partners. Same‐sex couples have been oversampled both in order to provide better information about the difficult‐to‐study sexual minority population, and in order to provide new perspectives on the changing nature of same‐sex couple mating in the US. Another key purpose is to examine how technology, specifically online dating and cell phone apps like Tinder and Grindr, affect relationship formation, relationship quality, attachment to the idea of monogamy, and relationship stability. The survey was conducted using sample from KnowledgePanel(spoken about in “HOW WAS THE DATA COLLECTED” section).
A few snapshots of the visualisation indicating status at different time intervals
A positive correlation can be derived from this data regarding how people meet each other, what does it say about the norms of the society then, how have people’s personalities been, what have people been spending most of their day-to-day time doing and how each of these things influence who/what type of counterpart they meet and the path that the relationship takes.
WHAT IS THE DATA?
The study is a national representative of American adults.
The survey data was collected from 3510 adults across the United States and 2862 of those had a spouse or main romantic partner. The rest 648 had responded to the survey based on past encounters and experiences.
Self-identified Lesbian, Gay, and Bisexual respondents were oversampled.
Because there were two branches to HCMST 2017 (for people with current partners and for people with past partners), variables that were consistent across branches were combined into new variables created by the Stanford research team which can be found at the tail of the dataset.
When a respondent refused to answer a question, the code “‐1” is used.
The data was collected for a survey by the Stanford University named “How Couples Meet and Stay Together(HCMST)” from the SSDS Social Science Data Collection Repository. The data has been divided between datasets for people’s responses along with their past partners and datasets for people’s responses with their current partners. A few noteworthy points are listed below:
HCMST 2017 surveyed a new cohort of subjects, building on the prior work around HCMST 2009. HCMST 2009 was generously supported by the National Science Foundation, grants SES-0751977 and SES-1153867.
Funding for HCMST 2017 comes from Stanford's United Parcel Service Endowment.
Surveys were performed by the online survey company GfK.
The data are nationally representative, as GfK recruits subjects into the panel by phone and by Address Based Sampling, and subjects without Internet access at home are given Internet access.
Follow-up surveys were implemented one and two years after the main survey, to study couple dissolution rates.
KnowledgePanel is the largest online panel that relies on probability‐based sampling techniques for recruitment; hence, the largest national sampling frame from which fully 11 representative samples can be generated to produce statistically valid inferences for study populations.
Participants completed the Fresh main survey in 11 minutes (median), and the Recontact survey in 1 minute (median), and overall, the study had a completion time of 10 minutes (median).
For the Fresh Survey, the survey consisted of general population adults, age 18+, English‐language survey‐takers residing in the United States, with an oversample of LGB adults; for the Recontact survey, the sample consisted of respondents from earlier waves of the How Couples Meet study.
The Fresh survey consisted of two stages: initial screening and the main survey with the study‐eligible respondents. To qualify for the main survey, a panel member provided consent to participate in the study, and reported either to have a sexual or romantic partner, or to have had one on the past.
The Survey FORM used for data collection can be found here.
WHAT QUESTIONS DO PEOPLE WANT TO ASK?
A few questions have been answered inline.
Does the way they meet affect/effect their relationship in any way?
Do traditional couples and nontraditional couples meet in the same way?
What kinds of couples are more likely to have met online?
Is the way they meet indicative of the societal norms of that age? Can we draw any generic/indicative conclusions about that epoch from this visualisation?
People had more trust in general, as the percentage of couples who got married wthin a span of 1 year is more in 1970s compared to the 2010s.
Have the most recent marriage cohorts met in the same way their parents and grandparents did?
How do the divorce rates of nontraditional couples(2010s) compare to the couple divorce rates of more traditional same-race heterosexual couples(1970s)?
What do we get to know about the personalities of people from this? Are they outgoing? Introverted? Commitment-issues? Orthodox?
The most common way to meet women/men was through people whom they already knew(Family/Friend/Co-worker) in 1970s, hence the trust factor on both their acquaintances and the new person is more.
The most common way in 2010s has been through Online Portals or in a Bar/restaurant which doesn’t cast much light on the trust factor. But it does say that, people are taking relationships more casually and easy-going than they were in 1970s.
The dots are colour coded right from after passing the First Met stage depending on what it’s destination level would be, i.e., Married(Blue), Live Together(Dark Green), Romantic(Light Green) and First Met(Yellow). It is more visually appealing and gives the user an idea of the percentage of each of the category points in each of the transient levels, at any duration of their relationship. The way they change the colour of points after evolving from the “First Met” stage also portrays their attention to detail.
Each of the points stay for some time on proportional basis in each of the transient levels. It spares the user some time in introspecting how long the points belonging to each destination category has spent in the intermediate levels. It depicts the speed or the trend with which couples headed for marriage or Live Together take their relationships forward.
They have indicated the percentile of couples in each of levels atop the visualisation which is a huge bonus for the users. Letting users compare merely by visual representation of the size of each cluster would have been cumbersome.
There are a few light/dark green-coloured points(Romantic/Live-Together levels, respectively) in the married section which gives a vague idea about the ratio of couples who get divorced and land back in the appropriate coloured stages. Though it isn't exactly indicative of the divorce rate, the effort in representing the divorce rates by retaining colours in upper-transient levels should be appreciated. This phenomenon is illustrated in the second screenshot of the Intoduction-What is this visualisation about? section.
The colours used in the visualisation are colour-blind friendly as illustrated below.
There should be a functionality to let users enter a specific relationship duration as it’d help researchers more in visualising trends on periodic levels than having to go through the entire timeline from the beginning.
Functionalities to move forward/backward a couple of months/years would have come handy.
There should have been another level named divorce or and indicator depicting the percentage of divorced couples which would help us further understand if the duration spent before wedding is indicative of divorce.
There could have been an indicator of which ways(Through friends/family, Bar, Public Place, etc.) have led to more divorces. Colour coding could have been visually effective as it would give us an idea of which all ways have sustained in the past.
The percentage of weddings which happened within 3 years of meeting each other has reduce to half from 60% in 1970 to 30% in 2010.
Correspondingly, the percentage of couples choosing to live together is almost similar as those who opt to marry within the same time period in 2010s, where as the percentage of Live Together had not even begun to be a trend in 1970s.
Interestingly, the percentage of couples who were still in the First Met stage 3 years since meeting is almost double compared to those in 1970s. This is surprising as regardless of the time period, 3 years is too long to still be in the “First Met” stage.
At the end of two years, the percentage of people who proceed to marriage and those who stay in a live-in relationship are almost similar. Post this timestamp, most couples proceed for marriage.
The trend shows that many couples who have been romantically involved with each other for more than 8 years in the 2010s proceed directly to marriage, skipping/spending a very short span of time in the intermediate “Living Together” step when they decide to take their relationships forward.
There’s still a 3% population which is in “First Met” stage after 13 years of meeting each other in 2010s. Introverted much?
College and Primary/Secondary Schools have been a significant way of meeting people till the early 2000s. Even then, Schools have always maintained a clear lead ahead of colleges, depicting the academic scene in the country. School education has been common, whereas Colleges/Universities have seen much fewer people.
Percentage of Blind dates have been continually decreasing since the 1960s, which is understandable as people get to rely more and more on electronice to meet a person, they stalk them a few times to gather information on all the alerting points about the person that the level of trust in blindly meeting and getting to know another person is no longer present.
School levels are constantly on the decrease since 1960s which could possibly mean that the importance of conventional education has been lowering. Note that this doesn’t necessarily mean that the person cannot be home-tutored/emerge accomplished.
Volunteering organisations have been a constant contributor in getting two people together, which is a good indicator that people have been considerate towards the less fortunate.
Church has been an eye-catchy factor in the visualisation whose contribution in this domain has been flaky at the beginning(1970) and end(2010), but more or less constant in between. But this need not say much about the religiousness of the general populace, as it is a generic belief that the millenials are much less religiously oriented compared to the elders. As the study participants consist of variety of sample categories such as young adults, elders, elders with current spouse and elders with ex-spouses, this doesn’t serve much towards any solid insight.
Family has managed to maintain it's ranking consitently in the top five across the years, which is plausible both in terms of comraderie and trust.
Though the percentage of couples who end up marrying is low in 2010s than in 1970s as depicted below, reports claim that divorce rate per 1000 couples has dropped from 4.8 to 3.2. This could be attributed to the fact that millenials are handling romance in their own ways as opposed to being bound by norms of the family and society. This means that, though they take longer to reach the "Marriage" state-of-mind due to their many insecurities, they put forward a bold step when they decide to do so.