Psychometric evaluation of the Chinese version of the febrile convulsion knowledge scale for parents/caregivers: translation and validation study

According to the research purpose, the researchers designed a general data questionnaire, including age, relationships with children, educational level, job, income, etc.

The scale was developed by Fatma Toksoz in 2023 [14] on the basis of the literature [15, 16], with a total of 8 items and 3 dimensions. Basic knowledge (1.2.3.4), first aid knowledge (5.6), and long-term effects (7.8). A 5-point Likert scale was used, and the correct items were scored from 1 to 5 (1 = “strongly disagree,” 2 = “disagree,” 3 = “neutral,” 4 = “agree,” and 5 = “strongly agree”). Questions(2.3.5.6.7.8) are reverse scored from 5 − 1. The total score on the scale ranges from 8 to 40, with lower scores indicating poorer knowledge of FC among family members.

In this study, Fatma Toksoz, the original author of the questionnaire, was contacted by email and authorized to translate the Febrile Convulsion Knowledge Scale for Parents/Caregivers [14]. The original scale was translated and cross-culturally adapted strictly following the Brislin translation model [17] to form the Chinese version of the Febrile Convulsion Knowledge Scale for Parents/Caregivers. The specific steps were as follows: (1) Translation: The original English scale was translated into Chinese versions (T1,T2 ) by two bilingual researchers whose mother tongue was Chinese on the research team. The research team discussed the two Chinese versions of the questionnaire, discussed and resolved the controversial points, and integrated the first draft of the Chinese version(T3). (2) Back translation: Two bilingual native English teachers who had never been exposed to the source questionnaire translated T3 into English versions (E1, E2), which were combined into English version ET3. Finally, E3 was sent to the original authors of the scale for comments, and then the Chinese version (C1) was formed after discussion and modification with members of the research team. (3) Cultural adjustment: In November 2023, the Chinese version of scale C1 was culturally adjusted through the Delphi letter consultation method. Six experts (3 clinical pediatric nursing experts and 3 nursing management experts), all of whom had associate senior professional titles and a bachelor’s degree or above, were invited to participate. The number of years worked by the experts was 24.67 ± 3.44, all of which were more than 15 years. Combined with clinical work experience and professional theory, the context, cultural adaptability, and language expression of the C1 items in the Chinese version of the scale were evaluated with reference to the original scale. Finally, the research team modified the questionnaire according to expert opinions and developed the Chinese version (C2).

The questionnaire is available in two forms: online and paper. Participants scan the QR code to answer the questions online via the Questionnaire Star platform. Informed consent was obtained from caregivers of children before the investigation, and an anonymous collection method and voluntary principles were adopted. The questionnaires were distributed to parents and caregivers of children who met the inclusion and exclusion criteria via a convenience sampling method. The questionnaire was completed independently by the participants. Prior to completing the questionnaire, participants were informed of the purpose and significance of the study, how to complete the questionnaire, and were instructed not to complete the questionnaire incorrectly, erroneously, or confusingly. After completing the questionnaire, the data were collected. A total of 410 parents and caregivers completed the questionnaire. Invalid questionnaires, such as those for which only one option was repeated for all questions or for which there was obvious logical confusion, were screened out, and 402 valid questionnaires were ultimately obtained, for an effective recovery rate of 98.04%. Two weeks later, 30 parents and caregivers randomly selected from the participants used the same questionnaire again to assess the retest reliability of the Chinese version [18].

The critical ratio method and correlation coefficient method were used to screen the items. (1) Critical ratio method: The 402 questionnaires were sorted according to the total score from high to low, T-test was used to compare whether the difference between the high group (the top 27%) and the low group (the bottom 27%) was statistically significant [19]. When the critical ratio was > 3.000, the discriminability of the item was high [19]. (2) Correlation coefficient method: The correlation coefficient between 8 items and the translated scale was calculated. Whether each item of the translated scale could be retained was judged by the item-total score correlation (r > 0.4) [20] and Cronbach’s alpha coefficient of the deleted item.

Reliability is an indicator of the accuracy or consistency of the response of the measuring instrument to the measurement result, reflecting the true degree of the measured characteristics. That is, whether the measurement tool is true and consistent for the measured object or the measured variable. The more consistent the test results, the smaller the error and the higher the reliability [21]. Cronbach’s alpha coefficient, split-half reliability, and test-retest reliability were used to evaluate the reliability of the scale. Cronbach’s alpha coefficient of the translated scale and each dimension was calculated, and 0.70 was taken as the acceptable standard of the reliability coefficient to evaluate the internal consistency of the scale [22]. The intraclass correlation coefficient (ICC) was used to evaluate the test-retest reliability of the scale. The 30 parents of the children were retested two weeks later using the translated scale and analyzed with the results of the first measurement to assess the temporal stability and consistency of the scale. To measure the stability and consistency of the scale over time.

Six experts were selected to score each item of the Chinese version of the Febrile Convulsion Knowledge Scale for Parents/Caregivers from the perspective of content importance. A 4-point Likert scale was used(1 to 4 corresponding to “irrelevant” to “very relevant”). The item-level content validity index (I-CVI) and scale-level content validity index (S-CVI) were calculated based on the expert rating results. The I-CVI refers to the proportion of the total number of experts who give 3 or 4 points to a project; the S-CVI is the mean of the I-CVI for all items.Generally, I-CVI ≥ 0.780 and S-CVI ≥ 0.900 are required [21].

The parents and caregivers of 402 children were randomly divided into two groups: one group (n = 150) for EFA and the other group (n = 252) for CFA. In EFA, we need to pay attention to the following points [23]: (1) Kaiser‒Meyer‒Olkin (KMO) value > 0.600; (2) the Bartlett sphericity test was significant (p < 0.05); (3) there was no obvious double loading of

the heading item; (4) title item load > 0.500; (5) cumulative variance interpretation > 50%. Principal component analysis (PCA) was used to rotate and delete items with factor loads < 0.500 by the varimax method to extract common factors with eigenvalues > 1 [24]. In CFA, the model fitting indices mainly include the chi-square degree of freedom(χ2/df), root-mean-square error of approximation (RMSEA), root mean square residual (RMR), comparative fit index (CFI), goodness-of-fit index (GFI), norm fit index (NFI), Tuck-Lewis index (TLI), and value added fit index (IFI). It is generally believed that χ2/df < 3.000, RMSEA < 0.080, RMR < 0.050, CFI, GFI, NFI, TLI, and IFI > 0.900 meet the model fitting requirements [25].