Republicans: Still Happy Campers

by Paul Taylor, Pew Research Center
October 23, 2008

Chin up, Republicans. Despite the imploding stock market, the looming recession, the unpopular president and the dismal political polls, there's very good news in the one realm of life that's always been a special sanctuary for you.

Personal happiness.

Some 37% of you say you're "very happy" with your lives, compared with just 25% of Democrats who feel the same way, according to a new nationwide Pew Research Center survey conducted from Oct. 3 through Oct. 19 -- a period that's witnessed a race to the bottom between John McCain's poll numbers and the public's 401(k) account balances.

This big GOP edge in life's happiness sweepstakes is not a new phenomenon. Ever since the General Social Survey¹ first asked the question in 1972, Republicans have always been happier than Democrats. But the current gap is among the largest on record.

What accounts for it? And how has it been able to withstand this fall's avalanche of unhappy political news for the GOP?

Well, Republicans are different from Democrats. How so? Let us count the ways.

• They have more money.²
• They have more friends.³
• They are more religious.⁴
• They are healthier.⁵
• They are more likely to be married.⁶
• They like their communities better.⁷
• They like their jobs more.⁸
• They are more satisfied with their family life.⁹
• They like the weather better.¹⁰
• They have fewer financial worries.¹¹
• They're more likely to see themselves doing better in life than their parents did.¹²
• They're more likely to feel that individuals - rather than outside forces - control their own success or failure.¹³
• They have more of what they most value in life. (No, it's not money).¹⁴

All of these differences are associated with the partisan happiness gap. But a word of caution: It can get very tricky trying to disentangle causes from effects. Even though happiness research has become a cottage industry in recent years, no one has yet been able to nail down all the factors that explain the persistent -- and growing -- Republican happiness advantage. No definitive answers will be available here, either. However, a close examination of data from three Pew Research Center surveys taken in 2008 may shed new light on this intriguing gap.

Let's start with some key demographic indicators. As is well known, Republicans, on average, have more money than Democrats. And -- all those lectures from parents, teachers and pastors to the contrary notwithstanding -- it turns out that money and happiness tend to go hand in hand, not just in this country but all over the world.¹⁵

In the latest Pew survey, some 35% of those with family incomes above $100,000 a year report being very happy, compared with just 27% of those with annual incomes of less than $50,000. (A June Pew survey had found an even wider happiness gap by income -- 46% of the $100,000-plus income group reported being very happy in that survey, compared with 30% of the under-$50,000 group -- perhaps suggesting that the recent stock market miseries have disproportionately intruded on the happiness of those in higher income brackets).

Pew surveys also find that Republicans have more money than Democrats -- on average, about $18,000 more a year in annual family income, according to a February survey. And when it comes to household wealth, the Republican advantage over Democrats is even greater.

Next up on the demographic check list is marriage. Republicans are more likely than Democrats to be married, and married adults are more likely than unmarried adults to report that they are very happy -- 34% vs. 24%.

But let's stop right here, because these correlations raise as many questions as they answer. Is it that happy people get married, or that married people get happy? Does money lead to happiness, or happiness to money? Or might some unrelated factor be creating correlations which don't actually have any causal connection?

This is where regression analysis can be helpful. It is a statistical technique that measures the independent importance of each of a number of variables in predicting a phenomenon of interest -- in this case, the likelihood that someone is very happy.

For the purposes of this analysis, we constructed two regression models from the Pew survey data. The first looks at respondents' party identification along with their basic demographic traits (age, gender, race, ethnicity, income, education, marital status). It then computes the odds that any single one of these traits will have a bearing on a person's likelihood of being very happy, once all of the other variables in the model have been held constant. (For a more thorough explanation, see Appendix).

In Model One, income has the greatest impact on predicting happiness: a high income person has a 16% greater chance than a low income person at being very happy, once all the other characteristics in the model have been held constant. Next on the list in this model is being a Republican rather than a Democrat - this increases someone's odds of being very happy by 13%, once all the demographic variables in this equation have been controlled. In other words, if one were to compare a theoretical Republican and a theoretical Democrat who had the identical age, ethnicity, race, gender, income, marital status and education level, the Republican would be 13% more likely than the Democrat to be very happy.

To be clear, this does not prove that being a Republican "causes" happiness. But it does demonstrate that, even after setting aside all the demographic traits that separate partisans of the two parties, a Republican is still a bit more likely than a Democrat to be very happy.

Our second regression model (Model Two) throws two additional variables into the equation -- church attendance and health status. Neither of these variables is a demographic trait per se, but both have been shown, in surveys taken over the years, to be correlated with happiness. And indeed, as Model Two demonstrates, they are more powerful than any demographic trait in predicting happiness. Someone who reports being in excellent/very good health has 34% better odds of being very happy than does someone who reports being in poor health, once all other variables in the equation are held constant. Likewise, someone who attends religious services weekly or more often has an 18% better chance of being very happy than someone who attends services seldom or never.

Adding these two factors to the regression model has the effect of reducing, but not eliminating, the independent impact of partisan identification on happiness -- from 13% in the first model to 7% in the second model.

Seven percent may not sound like much, but it means that someone's party identification has a greater direct bearing on that person's odds of being very happy than does the person's race, ethnicity or gender. In both models, none of those factors has any independent impact someone's chances for happiness, once the regression analysis has done its little magic. (To be sure, those traits may well have an indirect impact on happiness, because they correlate with variables that predict happiness, such as income, marriage and church attendance).

Where does this small but persistent GOP happiness edge come from? A February 2008 Pew Research Center survey looked at differences between Republicans and Democrats in their life priorities, values and experiences, and explored how these differences relate to overall levels of life satisfaction. The story that emerges from the survey data is quite consistent: advantage Republican.

Read the full report at pewsocialtrends.org

---

Notes

¹ A National Science Foundation-funded project that has conducted 26 nationwide surveys since 1972 that explore a range of demographic and attitudinal questions.
² Republicans have a median annual family income of $64,000; Democrats have a median annual family income of $46,000, according to a February, 2008 Pew survey.
³ Some 31% of Republicans and 26% of Democrats say they have a lot of friends in the community where they live. Also, 15% of Republicans and 26% of Democrats say they don't have many friends, according to an October, 2008 Pew survey.
⁴ Some 51% of Republicans attend religious services once a week or more; only 30% of Democrats do so, according to a February, 2008 Pew survey.
⁵ Some 37% of Republicans report their health condition is excellent, only 25% of Democrats say so, according to a February 2008 Pew survey.
⁶ Some 62% of Republicans and 44% of Democrats are currently married, while 7% of Republicans and 13% of Democrats are currently divorced, according to a February 2008 Pew survey.
⁷ Some 71% of Republicans rate their communities as "excellent" or "very good"; only 54% of Democrats say the same, according to an Oct ober, 2008 Pew survey.
⁸ Some 90% of Republicans and 86% of Democrats are completely or mostly satisfied with their job, according to a February 2008 Pew survey. This 4 percentage point difference falls short of statistical significance at the 95% confidence level.
⁹ Some 77% of Republicans and 62% of Democrats are "very satisfied" with their family life, according to a June 2008 Pew survey.
¹⁰ Some 27% of Republicans rate the climate where they live as "excellent" ; only 18% of Democrats think so, according to an October, 2008 Pew survey.
¹¹ For more details, see full report.
¹² Some 75% of Republicans and 61% of Democrats think that their own standard of living is better than the one their parents had when their parents were the age they are now, according to a February, 2008 Pew survey.
¹³ Some 69 % of Republicans and 58% of Democrats disagree with the statement that "Success in life is pretty much determined by forces outside our control," according to a February, 2008 Pew survey.
¹⁴ For more details, see full report.
¹⁵ Stevenson, Betsey and Wolfers, Justin, Economic Growth and Subjective Well-Being: Reassessing the Easterlin Paradox, April 16, 2008. www.nber.org