The relativity of relative satisfaction

Abstract

Freese [Evol. Hum. Behav. (2002)] claims that TV viewing has no effect on satisfaction with friendships relative to satisfaction with life in general. However, if one uses a seemingly better measure of general life satisfaction, most of my original findings remain or get even stronger. D 2002 Elsevier Science Inc. All rights reserved. Freese (2002) claims that TV viewing has no effect on satisfaction with friendships, relative to satisfaction with life in general, because all these measures of satisfaction with various aspects of life highly correlate with each other. I am not exactly sure what ‘‘satisfaction with friendships relative to overall life satisfaction’’ means. If TV viewing ceases to have a significant effect on such a relative measure of satisfaction, it seems to me that it is because watching TV increases satisfaction with life in general. Why should that be? What exactly does the concept ‘‘satisfaction with friendships relative to overall life satisfaction’’ measure? However, since Freese, in his critique, goes out of his way to take my empirical analysis on its own terms, despite his numerous skepticisms, I will return the favor and continue to use his notion of ‘‘satisfaction with friendships relative to satisfaction with life in general,’’ despite my reservations with the concept. If Freese wants to relativize satisfaction with friendships by overall life satisfaction, however, it appears that there is a better way of doing so in the GSS data. The GSS asks its respondents in every survey a specific question about satisfaction with life in general: ‘‘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?’’ It seems to 1090-5138/03/$ – see front matter D 2002 Elsevier Science Inc. All rights reserved. PII: S1090 -5138 (02 )00108 -3 E-mail address: satoshi.kanazawa@canterbury.ac.nz (S. Kanazawa). Evolution and Human Behavior 24 (2003) 71–73 me that this is a much more global measure of satisfaction or happiness with which Freese could have relativized satisfaction with friendships. As Table 1 shows, if I enter this measure (reversed coded as 1 = not too happy, 2 = pretty happy, 3 = very happy) in my original equations in order to control for the respondent’s overall life satisfaction, five out of my six original findings stand, and two of the three significant effects remain the same or get even stronger. The effect of watching TV news among men is no longer statistically significant (P= .1176); however, the effect of watching dramas and sitcoms for women is still statistically significant (P < .05) and the effect of watching PBS shows among men becomes even stronger (changing from P< .01 to P < .001). Note that, as Freese argues, the effect of overall life satisfaction on satisfaction with friendships is consistently and strongly significant (P < .0001 in most cases), indicating that the general and specific measures of satisfaction correlate highly, yet the significant effect of watching TV on satisfaction with friendships remains in two out of three cases. Table 1 The effects of TV friends on satisfaction with friendships relative to overall life satisfaction: US General Social Survey, 1993 TV friends Women Men Dramas and sitcoms 0.0859* (0.0422) 0.0278 (0.0539) TV news 0.0575 (0.0480) 0.1147 (0.0732) PBS shows 0.0371 (0.0403) 0.1697*** (0.0510)