Protective Behaviors. Among all respondents, individuals with personal experience engage in significantly more protective behaviors than those without personal experience (e.g., Table 16). Individuals who had coronavirus symptoms in the past seven days, tested positive for the virus, or live with someone who tested positive engage in, on average, 0.112 more disease-mitigating behaviors compared to those without personal experience. This is consistent with the difference in means insofar as individuals with personal experience have an average protective behaviors score of 4.370 while those without experience have an average score of 4.292. Examining the effects of partisanship on protective behaviors, I find that Republicans and Independents engage in significantly fewer protective behaviors than Democrats when looking at all respondents. Republicans’ take, on average, 0.543 fewer disease-mitigating actions those of Democrats and Independents engage in 0.188 fewer protective behaviors compared to Democrats. This makes sense when examining the mean disease-mitigating behaviors scores by party with an average number of protective behaviors being 4.024 for Republicans, 4.274 for Independents, and 4.621 for Democrats. Moreover, looking only at Republicans, having personal experience significantly increases protective behavior scores by 0.289 points (e.g., Table 18). Examining those with personal experience alone, I find that the partisan divide remains significantly pronounced with Republicans engaging in, on average, 0.363 fewer protective behaviors and Independents taking 0.272 fewer disease-mitigating steps compared to Democrats; however, the divide is diminished compared to the overall population (e.g., Table 17). Results for the other, more imperfect, measure of protective behaviors (i.e., number of places visited in the past 24 hours) reveal similar findings about the effects of partisanship. Republicans and Independents engage in significantly fewer disease-mitigating behaviors compared to Democrats when looking at all respondents. Republicans take, on average, 1.093 fewer protective actions than Democrats and Independents engage in 0.644 fewer disease- mitigating behaviors compared to Democrats. This is consistent with the difference in means insofar as Republicans’ average protective behaviors score is 8.917 while Independents’ mean score is 9.407 and Democrats’ is 10.012. Looking at those with personal experience, I find that the partisan divide nearly disappears. Republicans and Independents still engage in fewer protective behaviors than Democrats, but the difference is not statistically significant for Republicans and only mildly significant for Independents. Examining those without personal experience, however, I find that the partisan divide is even more pronounced than it is in the general population with Republicans engaging in, on average, 1.159 fewer protective behaviors than Democrats and Independents taking 0.595 fewer disease-mitigating steps compared to Democrats. Results from the three datasets are very much in line with what I would expect the effects of partisanship to be on risk perception, protective behaviors, and COVID-19 knowledge. The Qualtrics Questionnaire, ▇▇▇▇▇▇ Family Foundation Health Tracking Poll, and Gallup Panel each indicate that, among all respondents, Democrats are more likely to perceive COVID-19 as a threat compared to Republicans and Independents. Across a wide range of indicators – from perceived risk of contracting the virus and severity of illness to general anxiety about what the future holds and predictions about case numbers in the U.S. – Democrats perceive more personal vulnerability to the virus and express higher levels of concern about their own and others’ health. This is consistent with recent studies on divergent attitudes toward the coronavirus along party lines and longstanding trends in political polarization (▇▇▇▇▇▇▇; ▇▇▇▇▇▇▇▇ et al., 2020). As predicted, the partisan divide on risk perception is mitigated among healthcare workers and those with personal experience compared to the overall population. Within the healthcare worker and “personally experienced” cohorts, while Democrats are still more likely to indicate concern about contracting the virus compared to Republicans (and Independents in the case of the Kaiser Tracking Poll and Gallup Panel), the party difference largely diminishes; in fact, the divide becomes statistically insignificant in the Qualtrics Questionnaire. It makes sense then that, among all respondents in the questionnaire and panel, Democrats are also significantly more likely than Republicans and Independents to engage in protective behaviors ranging from purchasing a face mask and frequent hand washing to keeping away from crowded places and working from home. This partisan divide is mitigated, however, among healthcare workers and those with personal experience compared to the overall population as I expected. Democrats in both groups are still more likely to engage in disease-mitigating behaviors compared to Republicans and Independents, but the magnitude of difference is either diminished (i.e., Gallup) or no longer statistically meaningful (i.e., Qualtrics). There is strong confirmation from the datasets that, in general, partisanship is predictive of divergent coronavirus attitudes and behaviors, which become less polarized when examining healthcare workers and experienced individuals alone. What is more unclear, however, is whether or not healthcare workers and those with personal experience have increased risk perception and protective behaviors compared to non- healthcare workers and non-experienced individuals. Results from the Qualtrics dataset indicate that healthcare workers do report higher levels of concern about contracting the virus, but this result is inconsistent across measures of risk perception. While healthcare worker status is predictive of increased worry on my primary risk perception indicator (i.e., perceived risk of getting infected with coronavirus on a scale from 0–10) and one additional measure (i.e., perceived risk of catching coronavirus by flying), it has an insignificant effect on perceived severity, attitudes toward government response, and predictions about the number of future cases. Results from the Kaiser dataset add to this inconsistency in that healthcare workers are more likely to believe “the worst is yet to come” as opposed to “the worst is behind us” when asked about their feelings toward the coronavirus outbreak, but the difference is not statistically significant. Similarly, both the Kaiser poll and Gallup panel indicate that those with personal experience have higher risk perception scores compared to non-experienced individuals, but the difference is not statistically meaningful. However, despite not having more concern about the virus, respondents with personal experience engage in significantly more protective behaviors than those lacking experience. A possible explanation for this inconsistency is that those who had a coronavirus scare or know someone who tested positive do not necessarily believe the virus is any more severe than an average individual does, but they are more willing to engage in disease-mitigating behaviors to protect themselves and others knowing first-hand that the virus can be easily contracted. For healthcare workers, however, the counterintuitive results do not end at risk perception. Data from the Qualtrics Questionnaire reveals that healthcare workers actually engage in significantly fewer protective behaviors compared to non-healthcare workers. While examining the effects of healthcare worker status on risk perception and protective behaviors was not explicitly sought out as part of my analysis, this result is important in evaluating the strength of the mechanism proposed in my theoretical framework. That healthcare workers would be more likely to perceive COVID-19 as a threat and comply with public health recommendations was an assumption I made in order to explain a potential partisan mitigation among this cohort. Specifically, I suggested that individuals working in healthcare should have baseline scientific knowledge by virtue of their profession that uniquely equips them to dismiss contradictory elite, media, and in-group messaging, thereby shifting Republican healthcare workers toward the Democratic platform (i.e., high perceived risk and protective behaviors) and shrinking the partisan divide. However, this assumption is not consistent with findings from the datasets. Both the Qualtrics Questionnaire and ▇▇▇▇▇▇ Family Foundation Health Tracking Poll indicate that healthcare worker status has an insignificant effect on correct COVID-19 knowledge. Among all respondents, Democrats do score significantly higher than Republicans and Independents on assessments testing coronavirus knowledge (such as understanding of whether COVID-19 is a contagious disease or children are at high risk for complications from the virus). Moreover, the partisan divide is mitigated among healthcare workers and those with personal experience – Democrats are still more likely to have correct knowledge about coronavirus, but party differences become statistically insignificant. Moreover, those with personal experience score significantly higher on COVID-19 assessments compared to non- experienced individuals. However, contrary to what I expected, it is not the case that healthcare workers have higher levels of knowledge about the virus and are therefore better equipped to dismiss rhetoric downplaying the severity or even existence of the virus. This unanticipated finding could be due to the imperfectness of my healthcare worker status measures; phrasing of the healthcare employment question in both the Qualtrics and Kaiser datasets does not limit healthcare worker status to nurses or physicians. The measurement could reasonably include any individual working in a healthcare delivery setting, from an emergency room physician to a dining room staff member. Therefore, those who are coded as healthcare workers do not necessarily have formal scientific training and may be skewing the results. Nonetheless, if “healthcare workers” do not necessarily have more accurate scientific knowledge about COVID-19 and therefore higher risk perception and more protective behaviors compared to non-healthcare workers, what else might explain the mitigated effects of partisanship on coronavirus knowledge, attitudes, and behaviors among the healthcare worker cohort? It may be the case that there are significant differences in cue taking between healthcare workers and the general population. Perhaps those who work in a healthcare delivery setting receive their political information (specifically about the coronavirus) from their colleagues whereas most other Americans take cues from highly polarized political elites. This would not necessarily mean that those in healthcare are any more informed than the general population about best scientific practices or have higher risk perception and avoidance behavior, but simply that they are less polarized as a group because the cues from coworkers are more salient than those from political leaders. Alternatively, it may be the case that healthcare workers make a stronger distinction between professional and personal life than I expected, meaning that receiving accurate scientific information at work does not carry enough weight in and of itself to shift political attitudes formed at home.
Appears in 1 contract
Sources: Distribution Agreement
Protective Behaviors. Among The Qualtrics dataset reveals some counterintuitive results in that, among all respondents, individuals with personal experience healthcare workers actually engage in significantly more fewer protective behaviors than those without personal experience compared to non-healthcare workers (e.g., Table 163). Individuals who had coronavirus symptoms in However, examining the past seven dayseffects of partisanship, tested positive for the virusI find exactly what I would expect – Republicans and Independents are significantly less likely than Democrats to take disease-mitigating steps when looking at all respondents (e.g., or live with someone who tested positive Table 4). Republicans engage in, on average, 0.112 more disease-mitigating behaviors compared to those without personal experience. This is consistent with the difference in means insofar as individuals with personal experience have an average protective behaviors score of 4.370 while those without experience have an average score of 4.292. Examining the effects of partisanship on protective behaviors, I find that Republicans and Independents engage in significantly 0.154 fewer protective behaviors than Democrats when looking at all respondents. Republicans’ take, on average, 0.543 fewer disease-mitigating actions those of Democrats and Independents engage in 0.188 take 0.238 fewer protective behaviors compared to Democrats. This makes sense when examining the mean disease-mitigating behaviors scores by party with an average number of protective behaviors being 4.024 for Republicans, 4.274 for Independents, and 4.621 for Democrats. Moreover, looking only at Republicans, having personal experience significantly increases protective behavior scores by 0.289 points (e.g., Table 18). Examining those with personal experience alone, I find that the partisan divide remains significantly pronounced with Republicans engaging in, on average, 0.363 fewer protective behaviors and Independents taking 0.272 fewer disease-mitigating steps compared to Democrats; however, the divide is diminished compared to the overall population (e.g., Table 17). Results for the other, more imperfect, measure of protective behaviors (i.e., number of places visited in the past 24 hours) reveal similar findings about the effects of partisanship. Republicans and Independents engage in significantly fewer disease-mitigating behaviors compared to Democrats when looking at all respondents. Republicans take, on average, 1.093 fewer protective actions than Democrats and Independents engage in 0.644 fewer disease- mitigating behaviors measures compared to Democrats. This is consistent with the difference in means insofar as Republicans’ Democrats average number of protective behaviors score is 8.917 6.259 while Independents’ mean score Republicans is 9.407 6.043 and Democrats’ Independents is 10.0125.969. Looking at those with personal experiencehealthcare workers alone, I find that the partisan divide nearly disappearsthere is no significant effect of partisanship on protective behaviors (e.g., Table 8). Republicans and Independents still engage indicate engaging in fewer protective disease-mitigating behaviors than compared to Democrats, but the difference is not statistically significant for Republicans and only mildly significant for Independentsmeaningful. Examining those without personal experienceonly non-healthcare workers, however, I find that the partisan divide is even more pronounced in some ways than it is in the general population with (e.g., Table 9). On average, protective behavior scores are 0.152 points lower for Republicans engaging inand 0.249 points lower for Independents compared to Democrats. Among all respondents, contrary to what I would expect, healthcare worker status has an insignificant impact on correct knowledge about COVID-19 (e.g., Table 3). However, examining the effects of partisanship, it is clear that Republicans are significantly less likely than Democrats to answer questions about coronavirus correctly when looking at all respondents (e.g., Table 4). Republicans answer, on average, 1.159 0.145 fewer protective behaviors questions correctly than Democrats Democrats, which makes sense when examining mean COVID-19 knowledge scores by party with average correct answers being 11.984 for Republicans and 12.243 for Democrats. Looking at healthcare workers alone, there is no significant effect of partisanship on correct COVID-19 knowledge (e.g., Table 8). Independents taking 0.595 fewer disease-mitigating steps still have lower scores compared to Democrats. Results from the three datasets are very much in line with what I would expect the effects of partisanship to be on risk perception, protective behaviors, and COVID-19 knowledge. The Qualtrics Questionnaire, ▇▇▇▇▇▇ Family Foundation Health Tracking Poll, and Gallup Panel each indicate that, among all respondents, Democrats are more likely to perceive COVID-19 as a threat compared to Republicans and Independents. Across a wide range of indicators – from perceived risk of contracting the virus and severity of illness to general anxiety about what the future holds and predictions about case numbers in the U.S. – Democrats perceive more personal vulnerability to the virus and express higher levels of concern about their own and others’ health. This is consistent with recent studies on divergent attitudes toward the coronavirus along party lines and longstanding trends in political polarization (▇▇▇▇▇▇▇; ▇▇▇▇▇▇▇▇ et al., 2020). As predicted, the partisan divide on risk perception is mitigated among healthcare workers and those with personal experience compared to the overall population. Within the healthcare worker and “personally experienced” cohorts, while Democrats are still more likely to indicate concern about contracting the virus compared to Republicans (and Independents in the case of the Kaiser Tracking Poll and Gallup Panel), the party difference largely diminishes; in fact, the divide becomes statistically insignificant in the Qualtrics Questionnaire. It makes sense then that, among all respondents in the questionnaire and panel, Democrats are also significantly more likely than Republicans and Independents to engage in protective behaviors ranging from purchasing a face mask and frequent hand washing to keeping away from crowded places and working from home. This partisan divide is mitigated, however, among healthcare workers and those with personal experience compared to the overall population as I expected. Democrats in both groups are still more likely to engage in disease-mitigating behaviors compared to Republicans and Independents, but the magnitude of difference is either diminished (i.e., Gallup) or no longer statistically meaningful (i.e., Qualtrics). There is strong confirmation from the datasets that, in general, partisanship is predictive of divergent coronavirus attitudes and behaviors, which become less polarized when examining healthcare workers and experienced individuals alone. What is more unclear, however, is whether or not healthcare workers and those with personal experience have increased risk perception and protective behaviors compared to non- healthcare workers and non-experienced individuals. Results from the Qualtrics dataset indicate that healthcare workers do report higher levels of concern about contracting the virus, but this result is inconsistent across measures of risk perception. While healthcare worker status is predictive of increased worry on my primary risk perception indicator (i.e., perceived risk of getting infected with coronavirus on a scale from 0–10) and one additional measure (i.e., perceived risk of catching coronavirus by flying), it has an insignificant effect on perceived severity, attitudes toward government response, and predictions about the number of future cases. Results from the Kaiser dataset add to this inconsistency in that healthcare workers are more likely to believe “the worst is yet to come” as opposed to “the worst is behind us” when asked about their feelings toward the coronavirus outbreak, but the difference is not statistically significant. Similarly, both the Kaiser poll and Gallup panel indicate that those with personal experience have higher risk perception scores compared to non-experienced individuals, but the difference is not statistically meaningful. However, despite not having more concern about the virus, respondents with personal experience engage in significantly more protective behaviors than those lacking experience. A possible explanation for this inconsistency is that those who had a coronavirus scare or know someone who tested positive do not necessarily believe the virus is any more severe than an average individual does, but they are more willing to engage in diseaseExamining only non-mitigating behaviors to protect themselves and others knowing first-hand that the virus can be easily contracted. For healthcare workers, however, the counterintuitive results do not end at risk perception. Data from the Qualtrics Questionnaire reveals I find that healthcare workers actually engage in significantly fewer protective behaviors compared to non-healthcare workers. While examining the effects of healthcare worker status on risk perception and protective behaviors was not explicitly sought out as part of my analysis, this result is important in evaluating the strength of the mechanism proposed in my theoretical framework. That healthcare workers would be more likely to perceive COVID-19 as a threat and comply with public health recommendations was an assumption I made in order to explain a potential partisan mitigation among this cohort. Specifically, I suggested that individuals working in healthcare should have baseline scientific knowledge by virtue of their profession that uniquely equips them to dismiss contradictory elite, media, and in-group messaging, thereby shifting Republican healthcare workers toward the Democratic platform (i.e., high perceived risk and protective behaviors) and shrinking the partisan divide. However, this assumption is not consistent with findings from the datasets. Both the Qualtrics Questionnaire and ▇▇▇▇▇▇ Family Foundation Health Tracking Poll indicate that healthcare worker status has an insignificant effect on correct COVID-19 knowledge. Among all respondents, Democrats do score significantly higher than Republicans and Independents on assessments testing coronavirus knowledge (such as understanding of whether COVID-19 is a contagious disease or children are at high risk for complications from the virus). Moreover, the partisan divide is mitigated among healthcare workers and those still significantly pronounced with personal experience – Democrats are still more likely to have correct COVID-19 knowledge about coronavirus, but party differences become statistically insignificant. Moreover, those with personal experience score significantly higher on COVID-19 assessments scores being 0.171 points lower for Republicans compared to non- experienced individuals. HoweverDemocrats (e.g., contrary to what I expected, it is not the case that healthcare workers have higher levels of knowledge about the virus and are therefore better equipped to dismiss rhetoric downplaying the severity or even existence of the virus. This unanticipated finding could be due to the imperfectness of my healthcare worker status measures; phrasing of the healthcare employment question in both the Qualtrics and Kaiser datasets does not limit healthcare worker status to nurses or physicians. The measurement could reasonably include any individual working in a healthcare delivery setting, from an emergency room physician to a dining room staff member. Therefore, those who are coded as healthcare workers do not necessarily have formal scientific training and may be skewing the results. Nonetheless, if “healthcare workers” do not necessarily have more accurate scientific knowledge about COVID-19 and therefore higher risk perception and more protective behaviors compared to non-healthcare workers, what else might explain the mitigated effects of partisanship on coronavirus knowledge, attitudes, and behaviors among the healthcare worker cohort? It may be the case that there are significant differences in cue taking between healthcare workers and the general population. Perhaps those who work in a healthcare delivery setting receive their political information (specifically about the coronavirus) from their colleagues whereas most other Americans take cues from highly polarized political elites. This would not necessarily mean that those in healthcare are any more informed than the general population about best scientific practices or have higher risk perception and avoidance behavior, but simply that they are less polarized as a group because the cues from coworkers are more salient than those from political leaders. Alternatively, it may be the case that healthcare workers make a stronger distinction between professional and personal life than I expected, meaning that receiving accurate scientific information at work does not carry enough weight in and of itself to shift political attitudes formed at homeTable 9).
Appears in 1 contract
Sources: Distribution Agreement