Vaccine hesitancy has been identified as a growing phenomenon worldwide, with significant implications for human health.1,2 It has been defined as “a delay in acceptance, or refusal of vaccines, despite their availability” by the WHO’s Strategic Advisory Group of Experts on Immunization.3 In the human context, it is highly complex and exhibits substantial heterogeneity.1,2 That is, individuals not only differ in the degree of vaccine refusal but are also a highly diverse group in terms of such characteristics as political ideology, gender, or religiosity.4 A number of tools and strategies have been developed and implemented to reduce vaccine hesitancy and improve uptake.5 Among these, the parent attitudes about childhood vaccines (PACV) survey has been used by both researchers and practitioners to identify vaccine-hesitant parents and parental concerns and to improve communication between medical providers and patients.6–8 Importantly, such communication is a very low-cost strategy that avoids the political repercussions of other effective tools, like vaccination mandates.9,10 The PACV contains 15 items, is self-administered, and takes less than 5 minutes to complete.6,7 It has been adapted and translated to various contexts,11–13 and various shortened versions of it have also been developed.14 Tools like the PACV are commonly used in human medical settings, and here particularly in, and they are also important tools for researchers.8
Recent research indicates that vaccine hesitancy may not be confined to human vaccinations and may also be an increasing phenomenon among pet owners.15,16 While we know comparatively little about vaccine hesitancy among pet owners and how it affects their decisions to seek vaccinations for their pets, there are reasons to believe that spillover to pet vaccinations is taking place.15,17 Indeed, some initial evidence points to declining pet vaccination rates over time.18,19 Analogous to its human counterpart, vaccine hesitancy among pet owners is cause for alarm because of its implications for pet and public health16,18,19 and the impact on veterinarians, other care providers,16,18,19 and pet owners themselves.20,21 Gaining a better understanding of pet owners’ attitudes toward vaccination is crucial to better identify vaccine hesitancy as well as to develop and implement potential interventions in order to augment vaccine uptake. A standardized, easy-to-administer instrument to assess attitudes about vaccines among pet owners is essential in this regard. It seems reasonable that the PACV, which has been adapted to various contexts,11–13 may serve as a starting point in this process. The purpose of this study was to conduct an analysis of the validity and reliability of the PACV to American dog and cat owners and to assess suitability for identifying vaccine-hesitant pet owners.
Methods
The study received approval from the institutional review board at Texas A&M University. An online survey was created in Qualtrics and distributed via Lucid.
Survey distribution
The survey was conducted by a survey firm (Lucid), which utilized quota sampling to generate the study sample.22 Lucid was compensated at a rate of $1.50 per completed response. Lucid’s survey panel has been used extensively on issues related to health and social policy, including vaccinations23,24; has been validated; and has been identified as reliable.22,25 The survey was fielded from May 7 through May 15, 2024. Respondents became part of the survey via a double opt-in procedure, first opting to become a Lucid panel member and then agreeing to participate in the survey.22 A total of 10,038 respondents opted into the survey; 9,339 (93.0%) consented to take the survey, and 5,784 respondents completed the survey (61.9%). The completion rate was in line with recent research.15,23,24 In order to improve data quality, Lucid implemented an attention screener for all respondents prior to admittance to the survey. Some respondents who consented to take the survey were eliminated due to failure to pass 1 additional attention check (31.2%) or due to identification as potentially fraudulent using the survey software’s fraud detection tools (3.3%).26
Inclusion criteria
Respondents were included in the study if they indicated that they currently owned at least 1 dog or 1 cat.
Survey design
The PACV
The PACV contains 15 items, is self-administered, and takes less than 5 minutes to complete.6,7 The PACV items address respondents’ vaccination behavior (2 items), respondent concerns about vaccination safety and efficacy (4 items), and general attitudes and trust about vaccines (9 items). All items rely on a 3-point scale, a 5-point scale, or an 11-point scale. Each item is then rescored as 0 (“not hesitant”), 1 (“not sure or don’t know”), or 2 (“hesitant”), and results are summed. This sum is then further transformed into a 0-to-100 scale using linear transformation. Lastly, the measure is dichotomized, with scores equal or above 50 considered vaccine hesitant. The details of this process have been described extensively in the literature.6,7 Very limited adaptations were made to the 15-item PACV in order to make it applicable to dog and cat owners. This primarily consisted of replacing references to children with references to dogs and cats (see complete wording in Supplementary Figures S1 and S2). The 15-item PACV has previously been adapted to various contexts, such as different languages and countries,11–13,27 and several shortened 4- or 5-item versions of the measure have been developed.14,28 From a practical perspective, these shortened scales are valuable because they further reduce the burden of administration and increase completion rates.28 As a result, shortened versions of the PACV were developed and tested based on the procedures described in detail below.
Vaccination measures
After identifying dog and cat owners using a screening question, respondents were queried about 5 specific vaccinations in line with the author’s previous research.15 Both dog- and cat-owning respondents were asked about the vaccination status of their pets for rabies. Respondents who own more than 1 dog or cat were asked to answer all questions for their youngest dog or cat. They were also asked about 2 additional core vaccines.29 These core vaccines were canine parvovirus and canine distemper for dogs and feline panleukopenia and feline herpesvirus-1 for cats. Core vaccines are vaccines that are highly recommended by the veterinary community for all dogs and cats, independent of lifestyle or potential disease exposure, whereas recommendations for noncore vaccine are generally based on an individual assessment of risk.29 Pet owners were also asked about 2 noncore vaccines. These were canine influenza and Lyme disease for dog owners and feline chlamydia (chlamydophilosis) and feline Bordetella for cat owners.
Additional explanatory measures
In addition to the PACV, several commonly employed explanatory measures were utilized to predict vaccination status. To account for ideology,16,30 dichotomous measures for liberals and conservatives were derived from the standard survey questions on political ideology and then combining “extreme liberals” and “liberals” and “extreme conservatives” and “conservatives.” In addition, a dichotomous indicator variable for respondents who voted for President Trump in the last presidential election was included.30 Analyses also accounted for respondents’ religious activity (5-point scale).30 All respondents were also asked about their trust in health institutions and authorities, like veterinarians, the CDC, the NIH, and the FDA, each scored using 4-point measures, from “no confidence at all” to “a great deal.”9 Overall trust in these institutions and authorities, a 13-point index, was compiled by combining the 4 4-point questions about veterinarians, the CDC, the NIH, and the FDA.
Pet owners were asked about the number of pets at their residence and presented with 5 distinct options, from 1 dog or cat to 5 or more.20 Pet owners were also asked about the frequency of exposure of their pets to other pets outside their household, with options for (1) never, (2) 1 to 5 per month, and (3) more than 5 times per month.20 Respondents were also asked how long they had been with their current veterinarian, with response options for (1) less than 1 year, (2) 1 to 2 years, (3) 2 to 5 years, and (4) more than 5 years.
Lastly, standard control variables for pet owners’ gender,31 education,32 income,31 age,33 and race and ethnicity were included.5,34,35
Statistical analysis
In order to assess the PACV in the context of pet owners, a set of well-established methodological approaches were followed.6,11,28 The Bartlett test of sphericity was used to determine data appropriateness, and the Kaiser-Meyer-Olkin measure was used to determine sampling adequacy for exploratory factor analysis (EFA).28 Exploratory factor analysis was employed to determine the underlying structure of the PACV items as well as the relationship between them, including the identification of potential subscales.6,11,28 Mokken scale analysis (MSA; a procedure to assess whether survey items can be appropriately combined into scales),36 Cronbach α, and McDonald ω to determine reliability and internal consistency of the (sub)scales were conducted.13 The Mokken scale procedure was implemented using the Stata msp command.37 Bivariate analyses with each PACV item serving as the independent variable and vaccination status for each of the vaccines as the dependent variable were implemented.6 Lastly, construct validity was assessed by performing Fisher exact tests to assess the association between the PACV and vaccination status.6,11 A series of bivariate and multivariate linear probability models (LPMs) were estimated to assess predictors of vaccination status, including the PACV, retaining items with P < .05 in the reduced scales. Lastly, LPMs with the additional variables described above were estimated to assess correlates of vaccine hesitancy. Linear probability models were used to facilitate presentation as compared to traditional logit models, which indicated analogous results (omitted). Common procedures to assess the applicability of LPMs, such as tests for unusual and influential data and multicollinearity, were performed. All analyses were conducted in Stata/MP (StataCorp), version 18.0, for Mac.
Results
Demographic data were within several percentage points of national benchmarks for race, age, sex, income, and census region. Fit was further improved by weighting the data on gender, race, income, and education based on the current population survey (Supplementary Figure S3; Supplementary Table S1). Overall, 49.3% of respondents indicated that they owned at least 1 dog (N = 2,853), and 34.2% indicated that they owned at least 1 cat (N = 1,977; Supplementary Table S2). These numbers were in line with recent research15,16 as well as other national surveys.38,39 Among both dog and cat owners, 50.9% of respondents were women (Supplementary Table S2). The mean age for dog owners was 44.1 years and 45.9 years for cat owners. Lastly, dog owners (63.4%) and cat owners (69.7%) were predominantly White, followed by Hispanic (19.7% and 16.3%) and Black (9.8% and 7.5%)
Parent attitudes about childhood vaccines survey items
Table 1 (Supplementary Figure S1) displays data for each individual PACV item collapsed into the 3 categories of not hesitant (0), not sure/don’t know (1), or hesitant (2) for dog owners based on the transformation of the 3-point, 5-point, or 11-point scales described above. Several observations stand out. First, a large majority of respondents felt confident to discuss vaccines with their veterinarian (not hesitant: 91.0%), trusted their veterinarian (76.9%), and trusted the information they receive about vaccines (86.0%). Moreover, they also wanted a potential new puppy to receive all recommended vaccines (88.3%), had not avoided (73.6%) or delayed (72.2%) vaccinations unless their pet was allergic or sick, and believed that many vaccine-preventable diseases are severe (72.5%). Most dog owners also supported the recommended vaccine schedule (60.7%). Just over half of the respondents felt that it was not better for dogs to get immunity from getting sick than a vaccine (55.4%). However, a combined majority of respondents in the “not sure” and “hesitant” categories were concerned about vaccine safety (52.6%) and side effects (54.3%), as well as efficacy (50.5%), and thought that dogs were getting too many vaccines (58.7%). Importantly, a combined majority of respondents in the 2 categories also thought that it was better for dogs to get fewer vaccines when given simultaneously (86.6%). Lastly, per their own self-assessment (“Overall, how hesitant about dog vaccines would you consider yourself to be?”), most dog owners did not self-identify as vaccine hesitant (64.8% identified as not hesitant, 22.8% identified as hesitant, and 12.4% chose the unsure option).
Distribution of parent attitudes about childhood vaccines (PACV) survey scores and association with vaccination hesitancy among dog owners.
| Individual PACV score | ||||
|---|---|---|---|---|
| # | PACV item | Proportion not hesitant | Proportion don’t know | Proportion hesitant |
| 1 | Trust in dog’s veterinariana | 0.769 | 0.144 | 0.087 |
| 2 | Ability to discuss vaccine concerns with veterinariana | 0.910 | 0.067 | 0.023 |
| 3 | Trust in information on dog vaccinesa | 0.860 | 0.107 | 0.033 |
| 4 | Hesitancy toward dog vaccines | 0.648 | 0.124 | 0.228 |
| 5 | Vaccinate potential new puppya | 0.883 | 0.083 | 0.034 |
| 6 | Dog vaccine effectiveness | 0.495 | 0.169 | 0.336 |
| 7 | Dog vaccine safety | 0.478 | 0.164 | 0.357 |
| 8 | Dog vaccine side effects | 0.457 | 0.126 | 0.417 |
| 9 | Better to give fewer vaccines simultaneously | 0.134 | 0.399 | 0.467 |
| 10 | Immunity from illness better than vaccine | 0.554 | 0.246 | 0.200 |
| 11 | Many vaccine-preventable diseases are severe | 0.725 | 0.207 | 0.069 |
| 12 | Dogs get more vaccines than are good for them | 0.412 | 0.322 | 0.265 |
| 13 | Good to follow vaccine schedulea | 0.607 | 0.201 | 0.192 |
| 14 | Ever reject dog vaccine | 0.736 | 0.083 | 0.182 |
| 15 | Ever delayed dog vaccine | 0.722 | 0.078 | 0.200 |
Items included in reduced PACV scale.
Survey items scored as 0 (“not hesitant”), 1 (“not sure or don’t know”), or 2 (“hesitant) as described in the literature. Based on survey data collected by authors from an online survey of 5,784 US residents, May 7 to May 15, 2024. Of these, 2,853 owned a dog (49%). Weighted data displayed.
For cats (Table 2; Supplementary Figure S2), results were analogous. That is, many cat owners felt they could talk to their veterinarians about vaccines (not hesitant: 85.7%), trusted their veterinarians (73.7%), trusted information they receive about vaccines (82.0%), wanted a potential new kitten to receive the recommended vaccines (81.2%), had not delayed (67.4%) or rejected (69.0%) a vaccine other than for illness or allergy, thought many cat illnesses could be severe (67.2%), and thought the recommended vaccine schedule was a good idea (57.0%). However, only a slight majority thought that it is better to get the vaccine than to develop immunity from getting sick (55.9%). Again, a combined majority of respondents were concerned about efficacy (50.8%), safety (51.7%), and side effects (55.5%). Lastly, many thought that cats were getting too many vaccines (61.1%) and that it was better for cats to get fewer vaccines when given simultaneously (84.9%). Again, a majority (58.0%) did not self-identify (“Overall, how hesitant about cat vaccines would you consider yourself to be?”) as vaccine hesitant (24.3% identified as hesitant, and 17.7% chose the unsure option).
Distribution of PACV scores and association with vaccination hesitancy among cat owners.
| Individual PACV score | ||||
|---|---|---|---|---|
| # | PACV item | Proportion not hesitant | Proportion don’t know | Proportion not hesitant |
| 1 | Trust in cat’s veterinariana | 0.737 | 0.135 | 0.128 |
| 2 | Ability to discuss vaccine concerns with veterinariana | 0.857 | 0.111 | 0.032 |
| 3 | Trust in information on cat vaccinesa | 0.820 | 0.144 | 0.036 |
| 4 | Hesitancy toward cat vaccines | 0.580 | 0.177 | 0.243 |
| 5 | Vaccinate potential new kittena | 0.812 | 0.127 | 0.061 |
| 6 | Cat vaccine effectiveness | 0.493 | 0.215 | 0.293 |
| 7 | Cat vaccine safety | 0.483 | 0.176 | 0.341 |
| 8 | Cat vaccine side effects | 0.445 | 0.157 | 0.398 |
| 9 | Better to give fewer vaccines simultaneously | 0.151 | 0.424 | 0.425 |
| 10 | Immunity from illness better than vaccine | 0.559 | 0.259 | 0.183 |
| 11 | Many vaccine-preventable diseases are severe | 0.672 | 0.254 | 0.075 |
| 12 | Cat gets more vaccines than are good for them | 0.388 | 0.349 | 0.262 |
| 13 | Good to follow vaccine schedulea | 0.570 | 0.178 | 0.252 |
| 14 | Ever reject cat vaccine | 0.690 | 0.104 | 0.206 |
| 15 | Ever delayed cat vaccine | 0.674 | 0.099 | 0.228 |
Items included in reduced PACV scale.
Survey items scored as 0 (“not hesitant”), 1 (“not sure or don’t know”), or 2 (“hesitant) as described in the literature. Based on survey data collected by authors from an online survey of 5,784 US residents, May 7 to May 15, 2024. Of these, 1,977 owned a cat (34%). Weighted data displayed.
Exploratory factor analysis and item reduction
The Kaiser Meyer-Olkin test of sampling adequacy and the Bartlett test of sphericity indicated that factor analysis was appropriate for both cat and dog owners. With regard to dog owners, EFA identified 2 factors (ie, underlying dimensions) within the PACV, indicated by eigenvalues above 1 (Supplementary Figure S4). Factor 1 was made up of items 4, 6 through 10, 12, 14, and 15, whereas factor 2 was made up of items 1 through 3, 5, 11, and 13. However, item 11 (“I believe that many of the dog illnesses that vaccines prevent are severe”) fell below the recommended cutoff of 0.30. The lack of suitability of item 11 was further confirmed by MSA. As a result, 14 items (dog PACV-14) were retained. The monotonicity assumption (ie, the assumption that individuals with higher levels of the underlying trait have a higher probability of endorsing the item) was not rejected for this or any other analysis presented below. The PACVs can thus be considered to be an appropriate measure of the underlying trait.37,40 However, double monotonicity was not present.37,40 Factors were rotated using oblique (oblimin) rotation to improve interpretation (Table 3; Supplementary Figure S5). Subsequent MSA produced 2 subscales (scalability coefficients: scale 1, 0.421; scale 2, 0.461; Table 3). The standardized Cronbach α and McDonald ω for dog owners for all 14 items were 0.828 and 0.821. They were 0.829 and 0.825 for factor 1 and 0.726 and 0.728 for factor 2, indicating high reliability. Previous approaches to item reduction based on bivariate association between individual items and vaccination status were followed6 in the development of a reduced scale (Table 1). This approach retained items 1 through 3, 5, 13, and 14. Exploratory factor analysis on the reduced scale indicated 1 factor (Table 3). Mokken scale analysis indicated the elimination of item 14 and the identified 1 subscale (scalability coefficient, 0.458). The Cronbach α and McDonald ω were 0.726 and 0.728. The reduced scale (dog PACV-5) fulfilled monotonicity and double monotonicity assumptions.
Results for exploratory factor analysis and Mokken scale analysis for dog owners.
| Dog PACV-14 | Dog PACV-5 | |||||
|---|---|---|---|---|---|---|
| Factor loadings (> 0.3) | Loevinger H coefficients (> 0.3) | Factor loadings (> 0.3) | Loevinger H (> 0.3) | |||
| PACV item | Factor 1 | Factor 2 | Scale 1 | Scale 2 | Factor 1 | |
| 1 | 0.593 | 0.481 | 0.591 | 0.454 | ||
| 2 | 0.592 | 0.407 | 0.543 | 0.412 | ||
| 3 | 0.725 | 0.304 | 0.686 | 0.494 | ||
| 4 | 0.610 | 0.471 | ||||
| 5 | 0.466 | 0.445 | 0.509 | 0.401 | ||
| 6 | 0.716 | 0.454 | ||||
| 7 | 0.825 | 0.519 | ||||
| 8 | 0.736 | 0.482 | ||||
| 9 | 0.424 | 0.318 | ||||
| 10 | 0.534 | 0.387 | ||||
| 11 | ||||||
| 12 | 0.556 | 0.419 | ||||
| 13 | 0.473 | 0.474 | 0.550 | 0.505 | ||
| 14 | 0.408 | 0.382 | ||||
| 15 | 0.363 | 0.342 | ||||
| Scale | 0.421 | 0.461 | 0.458 | |||
Kaiser Meyer-Olkin test of sampling adequacy and the Barlett test of sphericity indicated that factor analysis was appropriate. Factors rotated using oblique (oblimin) rotation where appropriate. Based on survey data collected by authors from an online survey of 5,784 US residents, May 7 to May 15, 2024. Of these, 2,853 owned a dog (49%).
For cat owners, EFA again indicated 2 factors, items 4, 6 through 10, 12, 14, and 15 for factor 1 and items 1 through 3, 5, 11, and 13 for factor 2 (Table 4; Supplementary Figure S6). Factors were again rotated using oblique (oblimin) rotation to improve interpretation (Table 4; Supplementary Figure S7). Mokken scale analysis again produced 2 subscales (scalability coefficients, 0.410 and 0.551) and also indicated the removal of item 11 (Table 4). The monotonicity assumption was not rejected, whereas double monotonicity was rejected. The standardized Cronbach α and McDonald ω for cat owners for all 14 items were 0.853 and 0.849. They were 0.844 and 0.841 for factor 1 and 0.784 and 0.786 for factor 2, indicating high reliability (cat PACV-14). Based on bivariate association between individual items and vaccination status, a reduced scale containing items 1 through 5, 11, and 13 through 15 (Table 4) was created. Again, MSA indicated removal of item 11. The reduced 8-item scale (cat PACV-8) loaded on a single factor (ie, indicated that the items all measured the same underlying construct) and MSA indicated 1 scale (scalability coefficient, 0.392). The monotonicity assumption was not rejected, whereas double monotonicity was rejected. The standardized Cronbach α and McDonald ω were 0.778 and 0.795.
Results for exploratory factor analysis and Mokken scale analysis for cat owners.
| Cat PACV-14 | Cat PACV-8 | |||||
|---|---|---|---|---|---|---|
| Factor loadings (> 0.3) | Loevinger H coefficients (> 0.3) | Factor loadings (> 0.3) | Loevinger H (> 0.3) | |||
| PACV item | Factor 1 | Factor 2 | Scale 1 | Scale 2 | Factor 1 | Scale 1 |
| 1 | 0.681 | 0.551 | 0.619 | 0.370 | ||
| 2 | 0.672 | 0.551 | 0.522 | 0.377 | ||
| 3 | 0.733 | 0.387 | 0.691 | 0.470 | ||
| 4 | 0.620 | 0.478 | 0.551 | 0.387 | ||
| 5 | 0.497 | 0.429 | 0.610 | 0.433 | ||
| 6 | 0.748 | 0.442 | ||||
| 7 | 0.854 | 0.515 | ||||
| 8 | 0.752 | 0.476 | ||||
| 9 | 0.425 | 0.279 | ||||
| 10 | 0.565 | 0.400 | ||||
| 11 | ||||||
| 12 | 0.610 | 0.444 | ||||
| 13 | 0.547 | 0.317 | 0.645 | 0.421 | ||
| 14 | 0.410 | 0.392 | 0.489 | 0.373 | ||
| 15 | 0.318 | 0.321 | 0.451 | 0.348 | ||
| Scale | 0.410 | 0.551 | 0.392 | |||
Kaiser Meyer-Olkin test of sampling adequacy and the Barlett test of sphericity indicated that factor analysis was appropriate. Factors rotated using oblique (oblimin) rotation where appropriate. Based on survey data collected by authors from an online survey of 5,784 US residents, May 7 to May 15, 2024. Of these, 1,977 owned a cat (34%).
Based on the adjusted version of the 14-item versions of the PACV, 21.7% of dog owners (mean score on the 0-to-100 scale, 32.7) and 25.9% of cat owners (mean score of 33.6) were categorized as vaccine hesitant, whereas the shortened versions identified 8.2% (mean score of 14.0) and 15.3% (mean score of 21.6), respectively. The distribution of scores on the 0-to-100 scales can be found in Supplementary Figures S8–S11.
Association between PACV and vaccination status
Ninety-five percent of dog owners self-report vaccination against rabies, 89.9% against parvovirus, 88.6% against canine distemper, 83.5% against canine influenza, and 82.1% against Lyme disease (Supplementary Figure S12). Based on cat owners’ responses, 88.4% of cats were vaccinated against rabies, 77.0% against panleukopenia, 75.3% against herpes, 72.3% against chlamydia, and 71.6% against feline Bordetella (Supplementary Figure S12). These rates are in line with recent research.15
To assess construct validity, several approaches were utilized. The Fisher exact test indicated a significant association between vaccination status for each of the 5 vaccines and the dog PACV-14 (P < .001) and the dog PACV-5 (P < .001) as well as the cat PACV-14 (P < .001) and the cat PACV-8 (P < .001). A series of bivariate LPMs with vaccination status for the 10 different vaccines as the dependent variable and the various PACVs as independent variables were estimated. The bivariate models (Supplementary Tables S3 and S4) consistently found statistically significant negative associations between the various PACVs and vaccinations status, with vaccine-hesitant respondents exhibiting a reduced probability of vaccination by 0.060 to 0.108 for the dog PACV-14 (P < .002), 0.111 to 0.237 for the dog PACV-5 (P < .001), 0.112 to 0.160 for the cat PACV-14 (P < .001), and 0.251 to 0.320 for the cat PACV-8 (P < .001).
Lastly, a set of LPMs with additional covariates commonly used to assess vaccination status (Table 5) was estimated. For dog owners, identification as vaccine hesitant via the dog PACV-14 was associated with a reduction in the probability of vaccination from −0.103 to −0.057 (P < .001) and identification via the dog PACV-5 with a reduction from −0.209 to −0.100 (P < .001; Supplementary Table S5). Moreover, a relationship with a veterinarian longer than 1 year also substantially increased vaccination rates. Those who had been with their veterinarian for 1 to 2 years had an increased probability of vaccination of 0.054 to 0.134 (P < .020) compared to those who had been with their veterinarian for less than 1 year. For 2 to 5 years, the increase was 0.047 to 0.093 (P < .022), and for more than 5 years it was 0.036 to 0.089 (P < .035). Income was positively associated with vaccination for all vaccines (0.007 to 0.012; P < .012) except canine influenza and Lyme disease. Conservatives were less likely than moderates to vaccinate their dogs against canine influenza (−0.062; P < .006) and Lyme disease (−0.053; P < .020), whereas Trump voters were more likely to vaccinate against the latter (0.046; P < .023). No other variables were consistently significant. Results for the dog PACV-5 were analogous (Supplementary Table S5).
Predicted probability of questionnaire-assessed vaccine hesitancy (modeled as a dichotomous variable) based on dog owner reports of personal pet vaccination status using multivariate linear probability model analysis.
| Model | 1 | 2 | 3 | 4 | 5 |
|---|---|---|---|---|---|
| Variables | Rabies | Parvovirus | Canine distemper | Canine influenza | Lyme disease |
| Dog PACV-14 (vaccine hesitant) | −0.057 | −0.074 | −0.103 | −0.100 | −0.099 |
| (< .001) | (< .001) | (< .001) | (< .001) | (< .001) | |
| Dog PACV-5 (vaccine hesitant) | |||||
| Number of dogs in household: 2 | −0.014 | −0.035 | −0.006 | −0.004 | −0.040 |
| (.162) | (.039) | (.710) | (.858) | (.066) | |
| Number of dogs in household: 3 | −0.022 | −0.033 | 0.003 | −0.002 | 0.040 |
| (.319) | (.248) | (.908) | (.960) | (.207) | |
| Number of dogs in household: 4 | 0.001 | 0.018 | 0.024 | 0.037 | 0.048 |
| (.975) | (.569) | (.535) | (.451) | (.352) | |
| Number of dogs in household: 5 or more | 0.023 | 0.056 | 0.043 | 0.073 | 0.069 |
| (.240) | (.037) | (.334) | (.138) | (.215) | |
| Level of dog exposure: 1–5 times per month | −0.004 | 0.033 | 0.019 | 0.024 | 0.011 |
| (.710) | (.061) | (.280) | (.262) | (.613) | |
| Level of dog exposure: more than 5 times per month | 0.008 | 0.035 | 0.020 | 0.030 | 0.038 |
| (.488) | (.064) | (.344) | (.210) | (.128) | |
| Time with veterinarian: 1–2 years | 0.065 | 0.054 | 0.075 | 0.110 | 0.134 |
| (< 0.001) | (.019) | (.003) | (< 0.001) | (< 0.001) | |
| Time with veterinarian: 2–5 years | 0.047 | 0.057 | 0.068 | 0.068 | 0.093 |
| (.004) | (.012) | (.008) | (.021) | (.002) | |
| Time with veterinarian: more than 5 years | 0.036 | 0.063 | 0.060 | 0.089 | 0.078 |
| (.034) | (.006) | (.023) | (.002) | (.011) | |
| Female | −0.004 | 0.008 | −0.012 | −0.023 | −0.039 |
| (.603) | (.562) | (.397) | (.182) | (.031) | |
| Income (6 categories) | 0.007 | 0.012 | 0.012 | 0.006 | 0.001 |
| (.010) | (.003) | (.011) | (.273) | (.908) | |
| Education (4 categories) | −0.001 | −0.006 | −0.005 | 0.001 | −0.003 |
| (.897) | (.470) | (.556) | (.914) | (.751) | |
| Trust in health institutions (13 categories) | 0.002 | 0.006 | 0.005 | 0.006 | 0.004 |
| (.280) | (.032) | (.076) | (.079) | (.239) | |
| Religious activity (5–points) | −0.002 | −0.006 | 0.005 | 0.013 | 0.017 |
| (.433) | (.191) | (.386) | (.042) | (.011) | |
| Liberal | 0.002 | 0.014 | 0.003 | −0.009 | −0.035 |
| (.860) | (.411) | (.877) | (.688) | (.141) | |
| Conservative | 0.014 | −0.008 | −0.011 | −0.062 | −0.053 |
| (.179) | (.637) | (.573) | (.005) | (.019) | |
| Trump voter | 0.000 | 0.021 | 0.019 | 0.036 | 0.046 |
| (.998) | (.198) | (.302) | (.060) | (.022) | |
| White | 0.022 | 0.010 | 0.033 | 0.006 | 0.013 |
| (.474) | (.778) | (.372) | (.902) | (.768) | |
| Black | 0.027 | 0.008 | 0.006 | −0.010 | 0.024 |
| (.433) | (.843) | (.898) | (.843) | (.634) | |
| Asian | −0.002 | −0.003 | 0.002 | −0.007 | 0.001 |
| (.967) | (.946) | (.972) | (.907) | (.991) | |
| Hispanic | 0.044 | 0.041 | 0.026 | 0.020 | 0.009 |
| (.160) | (.276) | (.530) | (.690) | (.861) | |
| Age (in years) | 0.000 | −0.001 | −0.001 | −0.003 | −0.003 |
| (.098) | (.007) | (.166) | (< 0.001) | (< 0.001) | |
| Constant | 0.849 | 0.816 | 0.748 | 0.793 | 0.832 |
| (< .001) | (< .001) | (< .001) | (< .001) | (< .001) | |
| Observations | 2,667 | 2,667 | 2,667 | 2,667 | 2,667 |
| R2 | 0.041 | 0.042 | 0.038 | 0.054 | 0.059 |
Analyses based on survey data collected by authors from an online survey of 5,784 US residents, May 7 to May 15, 2024. Of these, 2,853 owned a dog (49%). P values in parentheses. Reference categories were level of cat exposure, never; number of cats in household, 1; time with veterinarian, less than 1 year; ideology, moderate; income, less than $14,000; education, less than high school. Coefficients indicate changes in the probability of vaccination. For binary variables, coefficients represent a change from 0 to 1. For ordinal variables, coefficients represent a discrete change by 1 unit. R2 can be interpreted as the difference between the average predicted probabilities in the 2 groups.
For cat vaccinations (Table 6), identification as vaccine hesitant via the cat PACV-14 (−0.137 to −0.092; P < .005) and the cat PACV-8 (– 0.255 to 0.213; P < .001; Supplementary Table S6) was associated with lower vaccination rates. For all vaccines, longer relationships with veterinarian again improved vaccine uptake (1 to 2 years: 0.138 to 0.167, P < .001; 2 to 5 years: 0.098 to 0.142, P < .002; more than 5 years: 0.078 to 0.137, P < .042). Except for rabies, income was associated with higher rates of vaccine uptake (0.019 to 0.025; P < .015). Trust in health institutions and authorities was positively associated with vaccinations (0.011 to 0.016; P < .011). Conservatives were less like to vaccinate against feline Bordetella (−0.068; P < .046). However, Trump voters were more likely vaccinate against herpes, chlamydia, and feline Bordetella (0.057 to 0.079; P < .047). Lastly, compared to no outside exposure, cats with 5 or more exposures per month exhibited decreased vaccination rates for rabies (−0.057; P < .027) and feline panleukopenia (−0.086; P < .014). Results for the cat PACV-8 were analogous (Supplementary Table S6).
Predicted probability of questionnaire-assessed vaccine hesitancy (modeled as a dichotomous variable) based on cat owner reports of personal pet vaccination status using multivariate linear probability model analysis.
| Model | 1 | 2 | 3 | 4 | 5 |
|---|---|---|---|---|---|
| Variables | Rabies | Panleukopenia | Herpes | Chlamydia | Feline Bordetella |
| Cat PACV-14 (vaccine hesitant) | −0.116 | −0.118 | −0.137 | −0.099 | −0.092 |
| (< .001) | (< .001) | (< .001) | (.001) | (.004) | |
| Cat PACV-8 (vaccine hesitant) | |||||
| Number of cats in household: 2 | −0.014 | −0.019 | −0.019 | −0.038 | −0.049 |
| (.474) | (.483) | (.488) | (.178) | (.094) | |
| Number of cats in household: 3 | 0.008 | −0.003 | −0.029 | −0.020 | −0.021 |
| (.764) | (.937) | (.509) | (.658) | (.642) | |
| Number of cats in household: 4 | 0.073 | 0.078 | 0.094 | 0.064 | 0.051 |
| (.023) | (.099) | (.058) | (.230) | (.369) | |
| Number of cats in household: 5 or more | −0.007 | −0.010 | 0.008 | −0.042 | −0.013 |
| (.873) | (.849) | (.880) | (.490) | (.819) | |
| Level of cat exposure: 1–5 times per month | −0.011 | −0.029 | −0.041 | −0.061 | −0.078 |
| (.660) | (.384) | (.227) | (.080) | (.032) | |
| Level of cat exposure: more than 5 times per month | −0.057 | −0.086 | −0.047 | −0.054 | −0.059 |
| (.026) | (.013) | (.159) | (.130) | (.101) | |
| Time with veterinarian: 1–2 years | 0.138 | 0.167 | 0.155 | 0.164 | 0.148 |
| (< 0.001) | (< 0.001) | (< 0.001) | (< 0.001) | (< 0.001) | |
| 3 time with veterinarian: 2–5 years | 0.098 | 0.141 | 0.142 | 0.123 | 0.128 |
| (< 0.001) | (< 0.001) | (< 0.001) | (.001) | (.001) | |
| Time with veterinarian: more than 5 years | 0.079 | 0.137 | 0.095 | 0.078 | 0.096 |
| (.007) | (< 0.001) | (.011) | (.041) | (.012) | |
| Female | −0.003 | −0.010 | −0.036 | −0.034 | −0.045 |
| (.840) | (.681) | (.126) | (.167) | (.071) | |
| Income (6 categories) | 0.008 | 0.020 | 0.025 | 0.019 | 0.022 |
| (.096) | (.007) | (.001) | (.014) | (.007) | |
| Education (4 categories) | −0.013 | −0.006 | −0.028 | −0.026 | −0.014 |
| (.221) | (.694) | (.051) | (.094) | (.373) | |
| Trust in health institutions (13 categories) | 0.011 | 0.011 | 0.014 | 0.016 | 0.014 |
| (.002) | (.010) | (.002) | (.001) | (.004) | |
| Religious activity (5-points) | 0.006 | 0.014 | 0.008 | 0.022 | 0.020 |
| (.346) | (.117) | (.331) | (.018) | (.033) | |
| Liberal | −0.018 | 0.003 | 0.018 | 0.015 | 0.009 |
| (.380) | (.915) | (.509) | (.619) | (.773) | |
| Conservative | −0.026 | −0.024 | −0.043 | −0.042 | −0.068 |
| (.316) | (.449) | (.177) | (.194) | (.046) | |
| Trump voter | −0.005 | 0.030 | 0.057 | 0.069 | 0.079 |
| (.822) | (.295) | (.046) | (.018) | (.009) | |
| White | 0.007 | 0.031 | 0.022 | 0.032 | 0.031 |
| (.874) | (.569) | (.675) | (.571) | (.591) | |
| Black | 0.054 | 0.027 | 0.068 | 0.049 | 0.098 |
| (.263) | (.716) | (.298) | (.496) | (.181) | |
| Asian | 0.056 | −0.003 | 0.002 | −0.007 | −0.014 |
| (.273) | (.968) | (.978) | (.932) | (.860) | |
| Hispanic | 0.082 | 0.090 | 0.071 | 0.093 | 0.056 |
| (.075) | (.137) | (.244) | (.139) | (.396) | |
| Age | −0.001 | −0.004 | −0.004 | −0.005 | −0.005 |
| (.273) | (< .001) | (< .001) | (< .001) | (< .001) | |
| Constant | 0.756 | 0.655 | 0.700 | 0.658 | 0.636 |
| (< .001) | (< .001) | (< .001) | (< .001) | (< .001) | |
| Observations | 1,827 | 1,827 | 1,827 | 1,827 | 1,827 |
| R2 | 0.089 | 0.090 | 0.099 | 0.094 | 0.085 |
Analyses based on survey data collected by authors from an online survey of 5,784 US residents, May 7 to May 15, 2024. Of these, 1,977 owned a cat (34%). P values in parentheses. Reference categories were level of cat exposure, never; number of cats in household, 1; time with veterinarian, less than 1 year; ideology, moderate; income, less than $14,000; education, less than high school. Coefficients indicate changes in the probability of vaccination. For binary variables, coefficients represent a change from 0 to 1. For ordinal variables, coefficients represent a discrete change by 1 unit. R2 can be interpreted as the difference between the average predicted probabilities in the 2 groups.
Discussion
Vaccination hesitancy is a growing public health threat.1,2 Recent research indicates that vaccination hesitancy is not confined to humans but also extends to pet owners, with the potential for reduced vaccination rates among pets.15,16 This article contributes to this budding field of research by assessing a common and widely applied survey tool, the PACV, to dog and cat owners. Overall, both standard and reduced versions of the PACV tool were explored, and both were determined to be reliable and valid measures of vaccine-hesitant pet owners. For both cats and dogs, 1 item (“I believe that many of the [dog/cat] illnesses that vaccines prevent are severe”) was removed from the tool. The shortened versions contained 5 identical items, with 3 additional items for the cat version (“Overall, how hesitant about cat vaccines would you consider yourself to be?,” “Have you ever decided not to have your cat get a vaccine for reasons other than illness or allergy?,” and “Have you ever delayed having your cat get a vaccine for reasons other than illness or allergy?”). All scales to indicate high reliability. Mokken scale analysis indicated satisfactory content validity and sufficient homogeneity among the different PACV items. Factor-loading structures for pet owners were different than in the human context, with 2 instead of 3 scales. Different scales are not uncommon and have also been identified in the human context.12,13 Reduced PACVs only exhibited 1 domain, indicating that the scale measured a single dimension. Construct validity was established as all 4 PACV measures were predictors of vaccination status. The reduced versions of the PACV served as better predictors of vaccination status than the full versions but identified fewer respondents as vaccine hesitant overall. However, the analyses using MSA in particular indicated some differences in the underlying vaccination hesitancy construct between dog and cat owners that deserve further exploration.
In terms of individual PACV items, a majority of respondents were concerned about the efficacy of vaccines as well as vaccine safety and side effects. In light of these findings, it is not surprising that many pet owners were concerned about giving their pets multiple vaccines at the same time and that a majority thought that pets were getting too many vaccines. At the same time, almost half of dog and cat owners felt it was better for their dog to get immunity from getting sick than the disease. Moreover, while a large majority did not self-identify as hesitant or generally delay or reject vaccines, substantial minorities did. Taken together, these findings raise concerns for public health going forward. Positively, a substantial majority trusted their veterinarians and felt they could discuss their concerns with them.
There are limitations to this study. First, this study is cross-sectional and provides an assessment at 1 point in time. Second, an internet-based nonprobability sample was utilized. While this approach is widespread and common in survey research today,22 it comes with certain limitations. Lucid has worked to improve data quality, and additional attention checks and fraud detectors were utilized to improve data and response quality. As for all survey research, analyses relied on respondents recall and truthfulness. However, it seems less likely today that social desirability bias may affect responses regarding vaccinations. Given the nature of this survey, it was infeasible to ask respondents to seek verification of vaccination status from their veterinarian. Hence, respondents may either not fully understand what vaccine their pet has received, hold misperceptions about what vaccinations are included in combination vaccines, or may not fully recall whether their pets are fully vaccinated. The limitations particularly apply to noncore vaccinations. In addition, the survey is nationally representative but does not speak to differences at subnational levels. It is also worth noting that the PACV does not directly assess the role that the cost of vaccinations may play in the delay or refusal of vaccinations. Lastly, not all scales achieved double monotonicity, indicating the potential for further refinement of survey items in the future.
Ultimately, the full and reduced versions of the PACV were valid and reliable tools to assess vaccination hesitancy in dog and cat owners. However, they categorized different percentages of respondents as vaccine hesitant and thus may be more or less useful depending on context, which may require further exploration. This is in line with the conceptualization of perceptions of vaccines along a continuum, from vaccine seeking to outright refusal.1,2 Future research should further seek to identify the utility of the different PACVs based on circumstances. More generally, future work should use confirmatory factor analyses to improve our understanding of vaccine-hesitant pet owners. As noted above, the various versions of the PACV served as correlates in the bivariate and multivariate models assessing vaccination status. These items may thus be useful for researchers to better identify factors associated with vaccine hesitancy in pet owners, to develop and implement evidence-based tools to increase vaccine uptake, and to track changes in pet owners’ attitudes toward vaccinations over time and in different geographies.
Supplementary Materials
Supplementary materials are posted online at the journal website: avmajournals.avma.org.
Acknowledgments
The author thanks the reviewers and editors for their helpful suggestions.
Disclosures
The author has nothing to disclose. No AI-assisted technologies were used in the generation of this manuscript.
Funding
The author has nothing to disclose.
ORCID
S. F. Haeder https://orcid.org/0000-0003-0077-6047
References
- 1.↑
Dubé E, Laberge C, Guay M, Bramadat P, Roy R, Bettinger JA. Vaccine hesitancy: an overview. Hum Vaccin Immunother. 2013;9(8):1763–1773. doi:10.4161/hv.24657
- 2.↑
Dubé E, Ward JK, Verger P, MacDonald NE. Vaccine hesitancy, acceptance, and anti-vaccination: trends and future prospects for public health. Annu Rev Public Health. 2021;42:175–191. doi:10.1146/annurev-publhealth-090419-102240
- 3.↑
SAGE Working Group. Report of the SAGE working group on vaccine hesitancy. World Health Organization. 2014. https://www.assetscienceinsociety.eu/sites/default/files/sage_working_group_revised_report_vaccine_hesitancy.pdf
- 4.↑
Lamot M, Kirbiš A, Vrdelja M. Exploring the inherent heterogeneity of vaccine hesitancy: a study of a childhood-vaccine-hesitant population. Vaccines (Basel). 2024;12(8):839. doi:10.3390/vaccines12080839
- 5.↑
Omer SB, Benjamin RM, Brewer NT, et al. Promoting COVID-19 vaccine acceptance: recommendations from the lancet commission on vaccine refusal, acceptance, and demand in the USA. Lancet. 2021;398(10317):2186–2192. doi:10.1016/S0140-6736(21)02507-1
- 6.↑
Opel DJ, Taylor JA, Mangione-Smith R, et al. Validity and reliability of a survey to identify vaccine-hesitant parents. Vaccine. 2011;29(38):6598–6605. doi:10.1016/j.vaccine.2011.06.115
- 7.↑
Opel DJ, Mangione-Smith R, Taylor JA, et al. Development of a survey to identify vaccine-hesitant parents: the parent attitudes about childhood vaccines survey. Hum Vaccin. 2011;7(4):419–425. doi:10.4161/hv.7.4.14120
- 8.↑
Opel DJ, Henrikson N, Lepere K, et al. Previsit screening for parental vaccine hesitancy: a cluster randomized trial. Pediatrics. 2019;144(5):e20190802. doi:10.1542/peds.2019-0802
- 9.↑
Haeder SF. Joining the herd? U.S. public opinion and vaccination requirements across educational settings during the COVID-19 pandemic. Vaccine. 2021;39(17):2375–2385. doi:10.1016/j.vaccine.2021.03.055
- 10.↑
Haeder SF. U.S. public support and opposition to vaccination mandates in K-12 education in light of the COVID-19 pandemic. Vaccine. 2023;41(48):7103–7115. doi:10.1016/j.vaccine.2023.10.016
- 11.↑
Truong BQQ, Ong KIC, Shibanuma A, Kiriya J, Jimba M. Adaptation and application of the parent attitudes about childhood vaccines survey tool in the Vietnamese language: a cross-sectional study. BMC Public Health. 2024;24(1):946.
- 12.↑
Abd Halim H, Abdul-Razak S, Md Yasin M, Isa MR. Validation study of the parent attitudes about childhood vaccines (PACV) questionnaire: the Malay version. Hum Vaccin Immunother. 2020;16(5):1040–1049. doi:10.1080/21645515.2019.1674112
- 13.↑
Olarewaju VO, Jafflin K, Deml MJ, et al. Application of the parent attitudes about childhood vaccines (PACV) survey in three national languages in Switzerland: exploratory factor analysis and Mokken scale analysis. Hum Vaccin Immunother. 2021;17(8):2652–2660. doi:10.1080/21645515.2021.1894894
- 14.↑
Opel DJ, Furniss A, Zhou C, et al. Parent attitudes towards childhood vaccines after the onset of SARS-CoV-2 in the United States. Acad Pediatr. 2022;22(8):1407–1413. doi:10.1016/j.acap.2022.06.014
- 15.↑
Haeder SF. Assessing vaccine hesitancy and support for vaccination requirements for pets and potential spillovers from humans. Vaccine. 2023;41(49):7322–7332. doi:10.1016/j.vaccine.2023.10.061
- 16.↑
Motta M, Motta G, Stecula D. Sick as a dog? The prevalence, politicization, and health policy consequences of canine vaccine hesitancy (CVH). Vaccine. 2023;41:5946–5950. doi:10.1016/j.vaccine.2023.08.059
- 17.↑
Lunz Trujillo K, Green J, Safarpour A, Lazer D, Lin J, Motta M. Covid-19 spillover effects onto general vaccine attitudes. Public Opin Q. 2024;88(1):97–122. doi:10.1093/poq/nfad059
- 18.↑
Kogan LR, Hellyer PW, Rishniw M. American and Canadian veterinarians’ perceptions on dog and cat core vaccination rates and the impact of the human medicine anti-vaxx movement on veterinary medicine. Can Vet J. 2021;62(3):247–252.
- 19.↑
Mattson K. Vaccine hesitancy: veterinary professionals face challenges surrounding vaccinations. AVMA. February 12, 2020. https://www.avma.org/javma-news/2020-03-01/vaccine-hesitancy
- 20.↑
Carlson D, Haeder SF, Jenkins-Smith H, Ripberger J, Silva C, Weimer DL. Monetizing bowser: a contingent valuation of the statistical value of dog life. J Benefit Cost Anal. 2020;11(1):131–149. doi:10.1017/bca.2019.33
- 21.↑
Weimer DL, Vining AR. Dog Economics: Perspectives on Our Canine Relationships. Cambridge University Press; 2024.
- 22.↑
Coppock A, McClellan OA. Validating the demographic, political, psychological, and experimental results obtained from a new source of online survey respondents. Res Polit. 2019;6(1):1–14. doi:10.1177/2053168018822174
- 23.↑
Haeder SF. Assessing parental intention to vaccinate against COVID-19, influenza, and RSV in the United States in late 2023. Vaccine. 2023;41(50):7503–7514. doi:10.1016/j.vaccine.2023.11.004
- 24.↑
Haeder SF. US seniors’ intention to vaccinate against RSV in fall and winter 2023–2024. Health Aff Sch. 2024;2(2):1–5. doi:10.1093/haschl/qxae003
- 25.↑
Stagnaro MN, Druckman J, Berinsky A, Arechar AA, Willer R, Rand D. Representativeness versus attentiveness: a comparison across nine online survey samples. PsyArXiv. 2024;22 February 2024 doi:10.31234/osf.io/h9j2d
- 26.↑
Ternovsk J, Orr L, Aronow PM, Kalla J. Evidence of rising rates of inattentiveness on Lucid in 2020. J Quant Descr Digit Media. 2022;2:1–35. https://osf.io/preprints/psyarxiv/h9j2d
- 27.↑
Olarewaju VO, Jafflin K, Deml MJ, et al. The youth attitudes about vaccines (YAV-5) scale: adapting the parent attitudes about childhood vaccines short scale for use with youth in German, French, and Italian in Switzerland, exploratory factor analysis and Mokken scaling analysis. Hum Vaccin Immunother. 2021;17(12):5183–5190. doi:10.1080/21645515.2021.1980314
- 28.↑
Oladejo O, Allen K, Amin A, Frew PM, Bednarczyk RA, Omer SB. Comparative analysis of the parent attitudes about childhood vaccines (PACV) short scale and the five categories of vaccine acceptance identified by gust et al. Vaccine. 2016;34(41):4964–4968. doi:10.1016/j.vaccine.2016.08.046
- 30.↑
Yadete T, Batra K, Netski DM, Antonio S, Patros MJ, Bester JC. Assessing acceptability of COVID-19 vaccine booster dose among adult Americans: a cross-sectional study. Vaccines (Basel). 2021;9(12):1424. doi:10.3390/vaccines9121424
- 31.↑
Callaghan T, Moghtaderi A, Lueck JA, et al. Correlates and disparities of intention to vaccinate against COVID-19. Soc Sci Med. 2021;272:113638. doi:10.1016/j.socscimed.2020.113638
- 32.↑
Wu YY, Zhang W. Demographic disparities in COVID-19 vaccine hesitancy among U.S. adults: analysis of household pulse survey data from Jul 21 to Oct 11 in 2021. Vaccine. 2023:7510–7514. doi:10.1016/j.vaccine.2022.10.094
- 33.↑
Campbell J, Kaur A, Gamino D, Benoit E, Amos B, Windsor L. Individual and structural determinants of COVID-19 vaccine uptake in a marginalized community in the United States. Vaccine. 2023;41(39):5706–5714. doi:10.1016/j.vaccine.2023.07.077
- 34.↑
Padamsee TJ, Bond RM, Dixon GN, et al. Changes in COVID-19 vaccine hesitancy among black and white individuals in the US. JAMA Netw Open. 2022;5(1):e2144470–e2144470. doi:10.1001/jamanetworkopen.2021.44470
- 35.↑
Khubchandani J, Macias Y. COVID-19 vaccination hesitancy in Hispanics and African-Americans: a review and recommendations for practice. Brain Behav Immun Health. 2021;15:100277. doi:10.1016/j.bbih.2021.100277
- 36.↑
Van Schuur WH. Mokken scale analysis: between the Guttman scale and parametric item response theory. Polit Anal. 2003;11(2):139–163. doi:10.1093/pan/mpg002
- 37.↑
Hardouin J-B, Bonnaud-Antignac A, Sébille V. Nonparametric item response theory using Stata. Stata J. 2011;11(1):30–51. doi:10.1177/1536867X1101100102
- 38.↑
YouGov. America’s Growing Dog Pack A Data-driven Profile of Dog Ownership in 2023. YouGov; 2023. Accessed April 11, 2024. https://commercial.yougov.com/rs/464-VHH-988/images/2023-05-YouGov-US-Dog-Owners-Report.pdf
- 39.↑
Brown A. About Half of U.S. Pet Owners Say Their Pets Are as Much a Part of Their Family as a Human Member. Pew Research Center; 2023.
- 40.↑
Wind SA. An instructional module on Mokken scale analysis. Educ Meas. 2017;36(2):50–66. doi:10.1111/emip.12153
