Because it is not always feasible to address a large number of beliefs in an intervention, it is important to determine which beliefs may be most important and are
likely to have the greatest impact on intention, if targeted. It was clear, and logical,
that the normative belief about one’s partner had far more impact than any other source
of influence on intention to use condoms with steady partners. Selecting the most
important behavioral and efficacy beliefs was not as simple, since multiple beliefs had
similarly high correlations with behavioral intention. After identifying beliefs significantly associated with intention, Fishbein and Cappella (2006) recommend using additional criteria suggested by Hornik and Woolf (1999) to select beliefs for intervention
targets. There should be enough people who do not already hold the belief to make intervention worthwhile, and it should be possible to change the belief through persua-
sive arguments.
To further evaluate beliefs based on the first of these criteria, the “behavioral beliefs” section of Table 4.3 includes percentages of intenders and non-intenders who
“strongly disagree” with each behavioral belief (these are all negative outcomes). The
“normative beliefs” and “efficacy beliefs” sections of the table include percentages
of intenders and non-intenders who “strongly agree” with each normative and efficacy belief. Less than 40 percent of non-intenders strongly disagreed with four behavioral beliefs:
show you think your partner is unclean
;
make partner think you don’t
love him/her
;
show partner you don’t trust him/her
; and
make partner think you are
having other partners.
These relatively low percentages, compared to intenders, make
these beliefs potentially important for intervention targets because of the large potential to increase percentages among non-intenders. Substantially higher percent disagreement by non-intenders, along with relatively small differences with intenders,
indicated that beliefs about embarrassment and lack of sexual release are probably
not useful targets for change. Similarly, five efficacy beliefs were endorsed by less
than 30 percent of non-intenders:
perception of ability to use condoms when carried
away
;
respondent had been drinking
;
partner had been drinking
;
using another method
of birth control
; and
partner doesn’t want to.
As with behavioral beliefs, there is con-
siderable room to change non-intender beliefs to be like intender beliefs on these five
efficacy issues. With respect to perceived norm, the greatest room for change is the
normative belief about one’s steady partner, with only 24 percent of non-intenders
believing that the steady partner wants to use condoms.
After identifying potential target beliefs, the next step is to consider whether each
of these beliefs can be changed through persuasive communications. The normative
belief about steady partners may be the most difficult to change through persuasive
arguments, unless the intervention also involves or targets the steady partner. The four
behavioral beliefs identified earlier appear to be reasonable targets for change through
persuasive arguments. For example, an argument may recommend that a person talk
to the steady partner about using condoms until they are both tested, and that this will
show enough love for the partner enough to protect him or her. This will help to counter
beliefs about showing you don’t love or trust the partner, thinking the partner is unclean,and making the partner think you are having other partners.
The five efficacy beliefs also seem reasonable to target through persuasive arguments that would provide strategies for a person to use to enhance the ability to use condoms in spite of drinking, being carried away, using another birth control method,and the partner being against using condoms