Using the General Social Survey data, I try to investigate whether general happiness is correlated with having children. For this purpose, I use the logit regression. Below are the variables used in the regressions.
The dependent variable :
HAPPY – Taken all together, how would you say things are these days – would you say that you are very happy, pretty happy, or not too happy? 1 = very happy, 2 = pretty happy, 3 = not too happy. Question asked from 1972 to 2010.
Since HAPPY is a dependent variable with 3 levels of response that is measured by the ordering of “very happy” down to “pretty happy” and “not too happy”, i.e. which means that HAPPY is not a continuous dependent variable, we should not use the ordinary least squares regression. Therefore I recoded this variable so that “very happy” and “pretty happy” take the value of 1 while “not too happy” takes the value of 0, as follows : HAPPY (d:1-2).
Independent variables :
CHILDS – Number of children. Recoded as follows : CHILDS (r:0;1-2;3-4;5-*). 4 point-scale.
REALINC – Family income on 1972-2006 surveys in constant dollars (base = 1986). Recoded as follows : REALINC (r:0-15000;15000-30000;30000-45000;45000-60000;60000-*). 5 point-scale.
DEGREE – Respondent’s highest degree. 0 = Less than high school, 1 = High school, 2 = Junior College, 3 = Bachelor, 4 = Graduate. 5 point-scale.
ATTEND – How often respondent attends religious services. 0 = Never, 8 = More than once week. Recoded as follows : ATTEND (r:0;1-2;3-4;5-6;7-8). 5 point-scale.
WORDSUM – Number words correct in vocabulary test. A proxy for IQ, correlation = 0.71; 0.83 for “g”. Recoded as follows : WORDSUM (r:0-2;3-4;5-6;7-8;9-10). 5 point-scale.
HEALTH – Condition of Health. 1 = Excellent, 2 = Good, 3 = Fair, 4 = Poor. 4 point-scale.
AGE (r:18-30;31-40;41-50;51-60;61-70;71-90). Respondent’s age. 6 point-scale.
YEAR (r:1972-1980;1981-1989;1990-1999;2000-*). Survey year. 4 point-scale.
Keep in mind that an independent variable with a low point-scale (say, 2) is expected to have a higher coefficient than an independent variable with a high point-scale (say, 10). In fact, a change in one unit of an independent variable with a high point-scale would have a very little effect, especially when the independent variable can take on many values (for instance, years, age, or income), on the dependent variable.
As for Selection Filter(s), I enter the following variables : RACE(1) SEX(2) MARITAL(1). In other words : white race, woman, married.
Then, for married white women, we have :
Allocation of cases (unweighted)
Valid cases – 4,197
Cases excluded by filters or weight – 41,547
Cases with invalid codes on variables in the analysis – 9,343
Total cases – 55,08
Now, I re-run a regression for non-married white women. Regarding Selection Filter(s), I enter the following variables : RACE(1) SEX(2) MARITAL(5). In other words : white race, woman, never married. Then, for non-married white women, we have :
Allocation of cases (unweighted)
Valid cases – 1,078
Cases excluded by filters or weight – 51,528
Cases with invalid codes on variables in the analysis – 2,481
Total cases – 55,087
Not surprising, here. HEALTH is the more important independent factor of happiness. And it is more important for non married women (-.843) than it is for married women (-.663). Also, being religious and rich account for happiness. Interestingly, being rich is a stronger factor of happiness among the non-married white women, since the coefficient is much higher for non-married women (.357 compared to .189 for married women). The other variables are not statistically significant, even at a 10% level of significance. And note that it is not money itself that makes us happy. The link between money and happiness is a mere illusion (Kahneman, 2006, pp. 5-8). Charles Murray also argues (2012, chapter 15) that some characteristics make us poor, and that it were these characteristics themselves, and not money, that would make us unhappy.
Now, regarding the number of children, we can see that having children is negatively associated with general happiness. A point worth noting is that this negative relationship is much stronger for non-married white women, which reflects the difficulty of being a single parent.
However, these figures are highly misleading. Indeed, it can be argued that the number of children decreases general happiness only when the number of children exceeds 2 or 3. Let’s see what we have with a cross-tabulation. For married women :
And for non-married women :
The data covers 1972-2010. But here, it is not clear that having one or two children makes us less happy than having no children. In contrast, having even one child substantially decreases general happiness among non-married white women. Again, this highlights the difficulty of raising a children without the assistance of a father. The pattern is very similar for non-married white men, and for married white men. Like I said before, feminism goes wrong.