my journey through the pursuit of love & happiness

Archive for the ‘duration-dependence model’ Category

This week in Quant I learned about duration-dependence/hazard models.

Duration-dependence models are when behavior change depends on how long a person has been exposed to a treatment variable.

For example, one might want to look at how long a person being on welfare affects them staying on welfare. Or, one may want to look at incarceration.

Of course the first thing that came to my mind was: sunk costs. Does duration-dependence affect people’s decisions to stay in a relationship?

So I emailed my professor who said, “I don’t know….I haven’t seen one, but I don’t know this literature that well. We do know that cohabiting couples in the US have very high rates of breakup relative to those in Europe…”

Having my curiosity left unfulfilled I used my fancy research assistant skills to look up studies on relations and the duration-dependence model.

There aren’t many.

The Dissolution of Intimate Relationships: A Hazard Model” by Diane Felmlee, et al seemed the best candidate to satisfy by curiosity.

To summarize the article’s findings:

Satisfying intimate relationships are crucial determinants of people’s happiness and general well-being (cited from a different study). But what they fail to do is define ‘intimate relationships.’ Are non-romantic relationships defined in this study as intimate relationships? This is something I will have to explore another time, however, if there is a study out there telling me that my happiness and well-being is all dependent on a fulfilling relationship with a man, than shoot me now.

Theories of relationship committment and stability suggest that the quality of a relationship affects whether or not a relationships continues. Beg and McQuinn (1986) found that love and maintenance behavior were significant predictors of relationship stability.

Additionally, the more an individual loves his or her partner AND engages in behavior to maintain the relationship the more likely it is that the relationship will remain intact.

The amount of time the couple spends together, the more closely their lives become intertwined and the more reasons they will have to remain together.

All in all, theĀ  amount of time spent together, when other variables are controlled (like love), is a good predictor of relationship stability.

Table 3 displays the multivariate estimates of the rate of break up following the first survey for current relationships. Prior duration in relationship is significant at p<.10. Sexual intimacy is not significant at any level. (I wonder if this means people are more likely to compromise on levels of sex satisfaction?)

The finding for relationship length implies that the longer a couple have been dating (before the first survey) the lower their rate of breaking up.

Additionally, the finding for the investment variable means that when other variables are controlled, the more an individuals believes that he/she has invested in a relationship the lower the rate at which the relationship is likely to end.

Other interesting findings about this study regard past relationships and how people look back on their fulfillment and reasons for break-up. This study, and other research suggests, that people look back on previous relationships in a way that maintains their self-esteem.


In addition to a few other factors a relationship is less likely to break up at any point in time the longer a couple has been dating before the first survey.

At the second survey prior duration in said relationship was no longer significant, however, the only variable that was was hours per week with partner at p<.05 level.

What does this really tell us about relationships and does it say anything about sunk costs in relationships?

I still believe that length of relationship should not be a deciding factor to stay in a relationship. If you are rationalizing staying with someone because you have been together for “so long” or “X months”, where X is considered long, than you should probably not be in that relationship.

I find it interesting that one of the significant variables is hours spent with your mate. I would like to see this study, with similar individuals, who are in long-distance relationships.

As I mentioned before, this study looks at social networks and their contribution to the dissolution of premarital relationships. It’s very interesting to see a study capture peer influences on the outcome of relationships. Not so surprising. Your mate either fits in with your social network or you build a new social network (your mate’s friends) in order to aid in the success of the relationship.

This makes me wonder how aware we are all of the reasons behind the dissolution of our romantic relationships.

If we are all aware of the determinants of breakups then we may be able to better understand ourselves, our wants and our needs in all of our relationships.

Or, at the very least, it makes for good food for thought.


Enter your email address to subscribe to this blog and receive notifications of new posts by email.

Join 2 other followers


July 2020
%d bloggers like this: