Friday, June 24, 2011

Debt boosts self-esteem??

There's an article on research from Ohio State University claiming that debt boosts the self esteem of young adults.  These people have clearly forgotten about the difference between correlation and causality. 
Researchers found that the more credit card and college loan debt held by young adults aged 18 to 27, the higher their self-esteem and the more they felt like they were in control of their lives.  The effect was strongest among those in the lowest economic class.

Only the oldest of those studied – those aged 28 to 34 – began showing signs of stress about the money they owed.
 Okay, they found a correlation between self esteem and debt.  They can't have found causality because they were examining surveys of students.  Researchers did not control who had debt and who didn't.  If they had taken a large group of students, measured their self esteem, randomly assigned half to receive debt and half to be a control group then they could have measured debt's effect on self confidence.  But as it is, all they know is that the two, debt and self confidence, go hand in hand.  They don't know why.

These researchers hypothesized that students would have a self esteem boost from debt because it was allowing them to accomplish great things in their lives like going to college through student loans or buying books on their credit cards.  Hold on a sec, how many college kids do you know where the majority of their credit card debt was because of textbooks?  It's unlikely and the researchers couldn't know what the students were buying.  So they came up with an alternate hypothesis:
“Obviously, they are probably using credit cards for multiple purposes.  Along with education spending, they could be using credit cards to pay for non-essential items.  They may feel good about their debt only because it allows them to buy the things they want without having to delay gratification.”
 That seems a bit more reasonable to me.  But there's a strong class factor in the correlation between self-esteem and debt:

Results showed that those in the bottom 25 percent in total family income got the largest boost from holding debt – the more debt they held, both education and credit card, the bigger the positive impact on their self-esteem and mastery.

Those in the middle class didn’t see any impact on their self-esteem and mastery by holding educational debt, perhaps because it is so common among their peers that it is seen as normal.  But they did see boosts from holding credit-card debt – the more debt, the more positive effects.

On the other hand, the study found that young adults who came from the most affluent families received no boost at all from holding debt.  “The wealthiest young people have the most resources and options available to them, so debt is not an issue for them,” Dwyer said.
I'm thinking that not only did these researchers jump to causality when all they had was correlation, but that they also jumped to the wrong causality.  My bet would be that debt doesn't cause higher self esteem, but that higher self esteem encourages students to take on more debt.  It would be a stronger correlation in the lower family incomes because these students have the least chance of paying off their debts.  You have to be ballsy to take on debt when it's more than your family's annual income.   You don't have to be confident or have high self esteem to take on debt if your family has a high income because you have a big safety net.

This explanation is a lot simpler than the researchers dual explanation of how debt improves self esteem and having three different reasons for the three different correlations by students' economic class.  Occam's razor, anyone? 

Do you think debt causes high self esteem or vice versa?  What do you think about the OSU study?

If you like this please Link Back to this article...

Related Posts by Categories


  1. Interesting article. I'm leaning toward your theory that it's the other way around. Higher self esteem (ballsy) people take on more debt. I wonder if this info can be used to screen loan applicants. The lower self esteemed folks, may be better at handling debt so they should be given the best rates. High self esteem folks, may be more of a risk. Similar to auto insurance screening for loans. Higher risk, higher premiums, higher rates, and higher qualifications.

  2. I'd think that it's pride that leads us to dig ourselves deeper into debt, giving a false since of "higher self esteem". In reality, it stinks!

  3. The pride idea was what I was thinking too. It doesn't really go into how they measure self esteem. The whole 'I deserve it because I'm me' mentality is what lands people in consumer debt issues to begin with.

  4. I just read the full study and your conclusions about their methods are way off. Not surprising as most people don't know much about how research is done. In any case, your concerns:

    1) Controlling for debt--they did this, why do you think they didn't? The survey was randomized, there were a huge number of respondents from different demographics and some had debt and some didn't, and in between. In this case the respondents are the children of a national 1979 survey, so they were easily able to control for the parents' demographics too.

    2. Correlation vs causality--their statistical methods are robust for correlation (ANOVA and chi-squared) AND causality (multiple regression). If you don't understand the math it can be confusing, but basically you compare different variables and factors in isolation and see how they affect each other. If you talk to the respondents at different points in time (as they did, these exact same respondents are tracked over time) you can see trends. In any case, YES, they did find causality, without a doubt. The effects were different for different demographic groups but the numbers are there. Sometimes research findings are surprising or unexpected...but that's why we do it.

    The authors intend to repeat the study to see how these same people feel about their lives as they get older:

    "The finding that negative effects of debt begin to emerge among the oldest respondents in
    the sample suggests it will be very important to continue to follow the experience of debt among this crucial cohort of young
    adults focusing not only on self-concept but also broader outcomes such as mental health, status attainment, and wealth

  5. We are wonder if this info can be used to screen loan applicants. The lower self esteemed folks, may be better at handling debt so they should be given the best rates.

  6. Weird- maybe being in debt is typical of young adults, and they pride themselves on being just like everyone else?

  7. Thank you for pointing out the it was correlation not causation. I have been seeing this study around and having people seem to have no idea about the difference has been driving me nuts. Again, thanks.

  8. I like your analysis of this study. We will never know the causality of debt and self-esteem, but I like your idea that perhaps the students with higher self-esteem and/or more confidence are more willing to do something "rash" like take on consumer debt. Sometimes, in money, a healthy dose of pessimism can be good.

  9. I can definitely see kids with higher self esteem taking on more debt. I feel like I tend to worry more and second guess myself on whether I'm making the right decisions, whereas a kid who thinks "of course I can handle paying off debt" is more willing to get into more of it.

    Then again, I also wonder if it makes kids feel more "grown up" to have their own credit card and being "paying" for their own college (although until your student loans are paid of, really department of education paid for your college...)