Measurement of the support for capital punishment has garnered much debate over the years. There are general measures of support for capital punishment and a variety of specific measures of support for capital punishment for specific crimes or when the alternative of LWOP is also presented in the question. Two measures of support for capital punishment were used in this study. On a 5-point scale, where 1 equals strongly agree and 5 equals strongly disagree, how much do you agree with the following statements:
1. General Measure: I favor the death penalty for people convicted of murder (reverse coded to consistently reflect high numbers support the death penalty and low numbers oppose the death penalty).
2. LWOP Measure: If I knew a murderer would stay in prison for the rest of their lives, I would not support the death penalty.
The General Measure of support for capital punishment was reverse coded, so that support for (agreement with) the death penalty increased with the numerical value of the response. The LWOP Measure was stated in the negative (‘‘I would not support the death penalty’’) so higher numerical values also indicate increased support for (agreement with) the death penalty. Because the literature shows that perceptions of the death penalty change quite dramatically based on the way the variable is measured, a separate analysis will be run for each operationalization of the variable. It is also possible that the relationship between the media and alternative measures of the dependent variable may be different because assessments of general support for the death penalty and support for the death penalty when the LWOP option is available may activate different cognitive processes for respondents when selecting a response. Table 2 shows the frequency distribution of each of these measures. Sixty-three percent of the respondents support (agree or strongly agree) the use of capital punishment with someone convicted of murder, compared to 47% of respondents who support the use of capital punishment when there is a LWOP option. A difference of means test finds these two measurements are statistically significant at p < .001 level (x ? 3.7 for General Capital Punishment Support Measure and x ? 3.2 for LWOP Measure). Because binary logistic regression is utilized in the analysis, both of these variables were dummy coded with strongly disagree, disagree, and neutral being coded as ‘‘0’’ and agree and strongly agree being coded as ‘‘1’’ and representing support for the death penalty.