Verbal consent was obtained from the participants. Ethical permission was obtained from Firat University Non-Interventional Research Ethics Committee (26.04.2022-8119).
Development of the Instrument:
Item Development
Identification of domain and item generation
As it is stated in the introduction part, public health researchers need to develop fluoride rejection screening tools, diagnostic tools, and evidence-based strategies for optimal preventive dental care decisions 13,14. However, such a tool about fluoride could not be found in the literature. Therefore, in order to bridge this gap and to make thisphenomena comparable with other population group and some other time, this study aims to develop a Fluoride Acceptance Scale (FAS).
In order to develop the Fluoride Acceptance scale, first a literature review was conducted 1-19. It was revealed that there was not any fluoride acceptance or rejection scales in the literature.
An item pool consisting of 15 items was created as a result of literature review. Both positive and negative expressions were used in the items. A 6-point Likert Scale ranging from 0 to 5 (strongly disagree:0, slightly agree:1, somewhat agree:2, quite agree:3, strongly agree:4, very strongly agree:5) was used.
Data Collection: The questionnaire was filled by 662 volunteers who applied to Firat University Faculty of Dentistry Hospital as patients and their companies on 2-6 May 2022. Inclusion criterias for the study are being between the ages of 18-65 and not having a problem that prevents communication. The questionnaire form used in the study include socio-demographic information form (gender, marital status, educational status, health worker, visiting dentist/per year, self-reported fluoride knowledge, permission for fluoride), the Fluoride Acceptance Scale, and the short form of the vaccine rejection scale developed by Kılıçarslan et al 19.
Data Analysis: The data were analyzed using SPSS Statistics Pack version 22.0 (IBM, Armonk, NY). A few missing values were tested with EM (expectation-maximization) Missing Value Analysis which is used to determine that the data are missing completely at random. Missing values are then replaced by imputed values. The missing data were completed with the most preferred (mode) values. Frequency, percentages, means and standard deviations were calculated for descriptive data. Statistical significance value was accepted as p <0.05.
Validity
Face Validity: To obtain Face Validity, the conceptual relationship of the scale with fluoride acceptance was evaluated by the authors, through asking their colleagues’ opinion and the pilot scheme.
Content validity: A total of 10 experts, 2 public health experts, 3 public health academicians, 3 dental acade-micians, 1 public dentist, 1 private dentist, were asked to evaluate the necessity, clarity and specificity of the items. Expert opinions were evaluated with Content Validity Index (CVI) Lawshe tecnique. For each item, the opinions of the experts as 4: Item is appropriate 3: Item should be slightly revised 2: Item should be reviewed seriously 1: Item is not appropriate, and their opinions were collected in the area they recommended to be edited 15.
Pretesting questions: The scale was piloted in a group of 50 people and feedback on the clarity and intelligibility of the items was received.
Sampling and survey administration: In scale development, it is recommended to reach participants equal to 5 or 10 times the number of items or to reach 200-300 participants 20,21. By using quota sampling methods, 662 volunteers over the age of 18 were reached from those who applied to Firat University Faculty of Dentistry as patients and their companies.
Factor Structure of the Scale: Barlett's Test of Sphericity, Kaiser-Mayer-Olkin were performed.
Convergent and Discriminant Validity: A negative correlation was found between the developed Fluoride Acceptance Scale and the Vaccine Rejection Scale.
Reliability
Internal Consistency: Cronbach's Alpha was used.
Split-half Reliability: The split-half reliability was performed by using Guttman's split-half coefficient.
Test-retest Reliability: To assess the test-retest reliability, the scale was administered to 30 participants twice, 15 days apart. Pearson's correlation coefficient was calculated for each item, and the factor structure obtained for the scale.