The vaccination breakdown effect provides evidence that the vaccination status of the source determines the rejection of calls for vaccination

Our sample (M= 49.09, South Dakota= 13.32, range [15; 83]) was representative of the average age of the adult population (M= 49.74), you(1169) = −1.68p= 0.094, D= − 0.05. However, we oversampled the women surveyed (57% against 52% in the population, z=3.43, pz= − 3.43, pz0.01, p > 0.999), but we undersampled partially vaccinated or recovered participants (29% versus 34% population, z= − 3.63, p z= 4.65, p

Confirmatory analyzes

We observed overall main effects consistent with the breakdown effect proposed by vaccination (correlations in Figure 1). Participants attributed less constructive motives, you(1167.50) = 4.80, pD= 0.28, 95% CI [0.17; 0.40]and less positive personality characteristics, you(1157.08) = 2.92, p= 0.004, D= 0.17, 95% CI[0.06; 0.29], to both vaccinated and unvaccinated commentators, despite identical message content. Overall, participants also reported feeling more threatened by calls to get vaccinated from the vaccinated source than from the unvaccinated source. you(1161.92) = −5.17, pD= − 0.30, 95% CI[− 0.42; − 0.19] (Table 1).

Figure 1

Correlation plot of dependent variables and condition.

Table 1 Mean ratings of comments and by source of comments (grouped by vaccination status of participants).

We analyzed participants’ behavioral planning and free-text counter-argument as behavioral indicators. Only 15.6% of respondents requested additional information and the source of the feedback did not have a significant direct impact on this measure, χ2(1, NOT=1170) p> 0.999. We expected criticisms from the vaccinated source to elicit more counter-arguments, as evidenced by a higher word count in the free-text response (assessed in LIWC201524). Previous research has observed that response length is strongly correlated with negative content when counter-arguing group reviews.25, and the word count thus provides a good behavioral indicator of vaccine flaw. Participants responding to the vaccinated commentator indeed used more words to express their opinion of the vaccine (Dnd= 22.50) than participants responding to the unvaccinated commentator (Dnd= 19.00), Wilcoxon signed rank tests O= − 3.00, 95% CI [− 5.00; − 1.00], p= 0.009, r= 0.08 (Table 2).

Table 2 Behavioral measures by source of feedback (grouped by vaccination status of participants).

Exploratory analyzes

Vaccination status of participants

Vaccination status may moderate the breakup effect as those who have already engaged in the behavior are less likely to distrust the message, although previous research indicates that even members of the commentator’s group are suspicious of cross-group criticism21. Indeed, participants’ vaccination status moderated the observed effects on message threat (Fig. 2b) and commentator ratings (Fig. 2c), as indicated by significant interactions between message source and message source. participant’s vaccination status, F(2, 1164) = 3.94, p= 0.008, ηp2= 0.01, and F(2, 1164) = 5.40, p= 0.001, ηp2= 0.01. This interaction was not significant for the comment pattern (Fig. 2a), F(2, 1164) = 1.46, p= 0.225, ηp2ps = 0.002 to ps = 0.014 to p= 0.049, but no threat, p= 0.444, or reviewer ratings, p= 0.162. Partially vaccinated participants did not show vaccine breakdown for the reason, p= 0.181, or reviewer ratings, p= 0.479, but for threat, p= 0.032. In sum, we observed consistent vaccination breakdown among unvaccinated or recovered people, but not those (partially) vaccinated.

Figure 2
Figure 2Figure 2

Boxplots with fiddle-plots (a) Evaluation of message motive by message source (unvaccinated vs. vaccinated) and vaccination status of participants. (b) Message threat by message source (unvaccinated vs. vaccinated source) and participant vaccination status. (vs) Ranking of message source by message source (unvaccinated vs. vaccinated source) and participant vaccination status. (D) Number of free-text response words by message source (unvaccinated vs. vaccinated source) and participant vaccination status. Eight extreme data points >300 words were omitted from the plot. To note.Box plot notches indicate 95% CIs.

Table 3 Mean feedback ratings by source of feedback for each vaccination status participant group.

The effects of requesting additional information or written responses (Fig. 2d) were not moderated by participants’ vaccination status, ps > 0.454 (group tests and effect sizes in Tables 4 and 5). However, ironically, participants who were already fully vaccinated (19%) were significantly more likely to ask for additional information than those who were not vaccinated (08%), χ2(1, NOT= 836) = 20.66, p2(1, NOT= 632) = 11.98, p= 0.001. Partially vaccinated participants were also significantly more likely to ask for additional information than unvaccinated ones (08%), χ2(1, NOT= 538) = 8.60, p= 0.003, or recovered (07%), χ2(1, NOT= 334) = 6.19, p= 0.013. No significant difference between partially and fully vaccinated participants, χ2(1, NOT= 634) = 0.34, p= 0.562, or recovered and unvaccinated participants emerged, χ2(1, NOT= 536) p= 0.945. Fully vaccinated participants provided shorter responses than recovered, ppp= 0.018, or unvaccinated participants, p= 0.034. The duration of response did not differ between fully and partially vaccinated participants, p= 0.200, neither participants cured and unvaccinated, p= 0.888. In total, the critical group of non-vaccinated people showed the strongest vaccine disruption effect (results detailed in appendix 1).

Table 4 Behavioral planning by source of feedback for each vaccination status participant group.
Table 5 Counter-argument by source of comments for each group of participants by vaccination status.

Structural Equation Modeling

We then explored the potential mechanisms underlying the vaccine disruption effect (Fig. 3; see Appendix 1, for detailed results). Source of feedback predicted constructiveness of assigned feedback, which predicted higher odds of participants engaging in behavioral planning as well as Aftercounter argument. The indirect effects of message source on planning behavior and counter-argument via motive were significant. Subsample analyzes (Figs. 4 and 5) showed that none of the mediators predicted planning or counterargument among fully vaccinated participants. Among partially vaccinated participants, an indirect serial effect as well as simple serial effects via message constructivity and threat emerged during planning but not in counterargument. Among retrieved participants, an indirect effect on behavioral planning via message motive emerged as well as on counter-argument via motive and threat. Finally, among unvaccinated participants, a serial indirect effect of message constructivity via commentators’ ratings on behavioral planning (but not counterargument) emerged. Apparently, for the unvaccinated, trust in the source is a particularly important determinant of the effects of vaccination on behavioral planning.

picture 3
picture 3

Results of analyzes of structural equation models. Dotted lines are insignificant paths. Estimates are standardized regression coefficients and ESs in parentheses.

Figure 4
number 4

Results of analyzes of structural equation models on behavioral planning according to vaccination status. Dotted lines represent insignificant paths. Estimates are standardized regression coefficients and ESs in parentheses; only significant paths are reported.

Figure 5
number 5

Results of analyzes of the structural equation model on the counter-argument by vaccination status. Dotted lines represent insignificant paths. Estimates are standardized regression coefficients and ESs in parentheses; only significant paths are reported.

Comments are closed.