Using the **General Social Survey** data, I try to investigate which factor is the most determinant of confidence in press and TV, as reported by GSS respondents. For this purpose, I use the logit regression. Below are the variables used in the regression.

CONTV – 1 = A geat deal, 2 = Only some, 3 = Hardly any. Recoded as follows : CONTV (d:2-3). This means that “only some” and “hardly any” take the value of 1 while “a great deal” takes the value of 0.

SEX – 1 = MALE, 2 = FEMALE.

DEGREE – 0 = Less than high school, 1 = High school, 2 = Junior College, 3 = Bachelor, 4 = Graduate.

WORDSUM – 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.

REALINC (r:0-15000;15000-30000;30000-45000;45000-60000;60000-*). 5 point-scale. Family income on 1972-2006 surveys in constant dollars (base = 1986).

POLVIEWS – 1 = Extremely liberal, 4 = Moderate, 7 = Extremely conservative. Recoded as follows : POLVIEWS (r:1-2;3;4;5;6-7). 5 point-scale.

ATTEND – 0 = Never, 8 = More than once week. Recoded as follows : ATTEND (r:0;1-2;3-4;5-6;7-8). 5 point-scale.

MARITAL – 1 = Married, 2 = Widowed, 3 = Divorced, 4 = Separated, 5 = Never Married. Recoded as follows : MARITAL (r:1;2-5). 2 point-scale.

AGE (r:18-30;31-40;41-50;51-60;61-70;71-90). 6 point-scale. Respondent’s age.

COHORT (r:1883-1910;1911-1930;1931-1950;1951-1970;1971-1992). 5 point-scale. Birth cohort of respondent.

Because the GSS asks the question in every survey year since 1973 we should also control for the survey year. COHORT may be used for this purpose.

Allocation of cases (unweighted)

Valid cases – 13,042

Cases excluded by filter or weight – 10,214

Cases with invalid codes on variables in the analysis – 31,831

Total cases – 55,087

Next, I run another regression, with a dependent variable labelled CONPRESS (“confidence in press”). Like CONTV, this variable needs to be recoded.

CONPRESS – 1 = A geat deal, 2 = Only some, 3 = Hardly any. Recoded as follows : CONPRESS (d:2-3). This means that “only some” and “hardly any” take the value of 1 while “a great deal” takes the value of 0.

Allocation of cases (unweighted)

Valid cases – 13,034

Cases excluded by filter or weight – 10,214

Cases with invalid codes on variables in the analysis – 31,839

Total cases – 55,087

It should be noted that while all coefficients are statistically significant (except REALINC in the second regression), the pseudo R-squared, which expresses the proportion of variance in the dependent variable explained by the entire set of independent variables, shows a very low value.

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.

Note that using YEAR variable – recoded as follows : YEAR (r:1972-1980;1981-1989;1990-1999;2000-*) – in place of COHORT does not change the result this much, for **CONTV** and **CONPRESS**.

As can be seen in the two regressions, the pattern is the same. Being a woman, educated, intelligent, rich, religious, conservative, married, and being old are negatively associated with the confidence in press and TV. The positive sign of COHORT shows that confidence in press and TV is decreasing over time.