Diet fat, salt and sugar as necessities of daily life, are important sources of essential macronutrients. However, excess intake of diet fat, salt and sugar is closely associated with the prevalence of non-communicable chronic diseases (NCDs) such as high blood pressure and cardiovascular diseases (CVDs)^1,2,3,4. The literature based on population, published in Lancet^5,6,7 demonstrated that decreased consumption of fiber-rich foods and increased consumption of high-fat, high-salt, high-sugar convenience foods and unhealthy lifestyle behaviors contribute to an increased risk of CVDs. This highlights the necessity for geographically targeted intervention strategies to ameliorate diet quality and mitigate the diet-related burden of CVDs.

For many industrialized countries, manufactured food and sugar-sweetened beverages are the primary sources of salt, saturated fat and sugar^8,9. Taxation of unhealthy or luxury food and drink items and product reformulation targeted at reducing salt, saturated fat, and sugar have proven effective^10,11,12. However, alternative strategies are needed for low-middle-income countries, such as China, where sodium and fat/oil intake are mainly from food prepared at home¹³. Population-based sodium and fat/oil reduction have emerged as effective strategies for managing hypertension, cardiovascular diseases, and obesity^13,14.

College canteens are places for centralized dining. One crucial stage in developing healthy personal habits is during higher education, where the diets of college students who have left their family environment also undergo changes that can impact their health in the short and long term. However, the intake of oil, salt and sugar is related to the individual’s eating behavior and the canteen staff’s health literacy and cooking behavior, which can supply a healthy eating environment. Individual behavior change is more likely to be facilitated and sustained if the environment within which choices are made supports healthful food options¹⁵. Therefore, evaluating the KAP of canteen staff about oil, salt and sugar is very important.

To our knowledge, no questionnaire assesses canteen staff’s KAP of “oil, salt, and sugar”. Thus, the primary aim of this study was to establish and validate a questionnaire to evaluate the KAP of canteen staff about “oil, salt, and sugar”. In this initial study, we hypothesized that the KAP questionnaire would have good measurement properties (internal reliability, test–retest reliability and validity). As part of the nutrition and healthy canteen evaluation system, this questionnaire will help assess the KAP of canteen staff and guide future intervention studies and nutritional and healthy canteen construction.

Phase 1: Questionnaire development

The questionnaire was developed by reviewing the literature and group discussions. We comprehensively reviewed relevant government documents and the published literature to identify KAP questionnaire items about oil, salt, sugar, and health. After eight rounds of group discussions, the final questionnaire comprised four sections containing 76 items.

Phase 2: Questionnaire validation

According to Walter¹⁶ and Mundfrom¹⁷, a sample size of at least 100 participants was considered adequate for test–retest reliability assessment (correlation coefficient) and factor analysis. Firstly, we randomly selected three college canteens in Shandong Province, and then 100 canteen staff as participants were selected in three different college canteens through cluster sampling. The inclusion criteria were adults aged ≥ 18 years and working experience ≥ 1 year who all signed an informed consent form and voluntarily participated in this study. Participants completed the paper questionnaire distributed by the researcher on-site and retested two weeks later. The questionnaire’s reliability and validity can be evaluated through the survey data.

Statistical analysis

Statistical analysis was performed using SPSS Statistics software (version 24.0). Mean and standard deviations were calculated for continuous variables, and frequencies and percentages were calculated for categorical variables. Cronbach’s α was used to evaluate the internal consistency of the questionnaires, which was 0.7 or higher, indicating a satisfactory level¹⁸. Pearson’s correlation coefficient was used to calculate the test–retest reliability. As per Cade et al.,¹⁹ correlation coefficients were defined as 0.5 or greater, which indicated higher reliability. Exploratory factor analysis using the principal axis factoring method and varimax rotation determined the construct validity.

The Kaiser–Mayer–Olkin (KMO) measure was used to assess sample adequacy, and values of more than 0.5 showed that the data was suitable for factor analysis. We applied the principal component method and varimax rotation to extract items into factors. The maximum variance orthogonal rotation method is used, and the principle of eigenvalue > 1 is selected for extraction. A two-sided p-value < 0.05 was considered statistically significant.

Institutional Review Board Statement

The current study was conducted according to the guidelines in the Declaration of Helsinki, and all procedures involving research study participants were approved by the Ethics Committee of the School of Public Health, Shandong University, reference number LL20200801.

Informed consent

All subjects gave their written informed consent for inclusion before participating in the study.

Following the systematic methodology, the final version of the knowledge, attitude, and practices (KAP) questionnaire was comprised of four main sections, a total of 76 items: demography (13 items), knowledge (21 items), attitude (20 items), and practice (22 items) (Table 1).

Socio-demographic profile of study participant

A total of 100 canteen staff participated in this study, with a valid response rate of 100%. Table 2 described the demographic characteristics of the study population. Most participants (66%) were females with a mean age of 40.3 ± 10.5 years old, residing in Rural areas (88%). Most participants (26%) had at least three years of work experience. The percentage of people’s monthly income over 3500 was 47%. A higher proportion of participants’ education was in Junior high school or higher (93%).

Reliability of the questionnaire

The reliability of the questionnaire was evaluated using Cronbach’s α coefficients for internal consistency and Pearson’s correlation coefficient for test–retest reliability. As shown in Table 3, the Cronbach’s α coefficients of the total questionnaire and Knowledge, Attitude and Practice dimensions, respectively, were 0.822, 0.830, 0.752 and 0.700. The Cronbach’s α coefficients were acceptable for knowledge, attitude and practice. Supplementary Table 1 showed the Pearson’s correlation coefficient for the test–retest reliability of the questionnaire. Correlation coefficients between the two administrations of 0.5 to 0.7 were common, and more than 0.7 can be indicated as very good reliability¹⁹. Based on Supplementary Table 1, coefficients for the knowledge, attitudes and practice scale were more significant than 0.586 (range: 0.586–1.000), 0.798 (range: 0.798–1.000) and 0.687 (range: 0.687–1.000), respectively, which indicated that each scale of the items had higher correlation. Overall, the correlation coefficient to the questionnaire was 0.969, which suggested that the questionnaire had good reliability.

Validity of the questionnaire

The questionnaire’s content and face validity were established through a satisfactory level of agreement among panelists. The construct validity was established by using exploratory factor analysis. The correlation matrix was used to assess the degree of correlation. The KMO test (0.656) and the Bartlett test of sphericity (Chi-squared, df = 378; P < 0.001) (Supplementary Table 2) showed that the data met the criteria required for factor analysis.

As Tables 4 and 5 showed, nine potential factors were identified, with a total of 26 items, which explained 70.923% of the variance. Factor 1 contained eight items on the practice of “oil, salt and sugar”; Factor 2 had three items on the knowledge of “oil, salt and sugar” and health; Factor 3 and Factor 4 both contained three items on the attitudes of “oil, salt and sugar” and health; Factor 5 included two items on attitudes towards the diet of low-sugar and Factor 6 contained two items on practice towards fatty foods and added sugar foods. Factor 7 had two items on salt-related knowledge and attitudes towards “oil, salt and sugar” reductions. Factor 8 contained two items on practice towards a healthy diet, and factor 9 included an item on behavior towards cooking with animal fat.

To date, there is no validated KAP questionnaire about the “oil, salt and sugar” used among canteen staff. Therefore, to fill this gap, we established and validated a questionnaire to evaluate the KAP of canteen staff about the “oil, salt and sugar”. This questionnaire is meant to be used by researchers interested in assessing these aspects in the Chinese canteen staff. It is easy to administer and only takes 15 to 20 min to complete for canteen staff.

The final questionnaire comprised three domains (knowledge, attitudes and practice) and 76 items. Canteen staff as participants (N = 100) randomly selected in three college canteens in Shandong Province finished the questionnaire twice to assess reliability and validity. The selected knowledge, attitude and practices scales showed acceptable and satisfactory internal consistencies and good reliability.

Reliability refers to the degree of consistency with which the same results are obtained when the same indicators or measurement tools are used to measure the same thing repeatedly. A questionnaire can be reliable but invalid, but a valid questionnaire is always reliable. Validity is the degree to which a questionnaire evaluates what it is intended to evaluate.

Cronbach’s α coefficient is often used to evaluate the internal consistency of the questionnaire, which is the method employed in most validation studies²⁰. Cronbach’s α value higher than 0.67 is almost acceptable, according to Paiva and Sasaki²¹. Values > 0.80 are considered good results for authors like Paiva²⁰ or excellent, according to Castro²². In our study, although five values were slightly under 0.7 for one construct, Cronbach’s α coefficients of the total questionnaire and each dimension were 0.822, 0.830, 0.752, and 0.700, respectively, which indicated the questionnaire as a whole had adequate internal uniformity and was at a satisfactory level. It was better than a questionnaire about the health literacy scale for low salt consumption²³. The Pearson’s correlation coefficient of the total questionnaire was 0.968, indicating that the questionnaire had high stability and good test–retest reliability¹⁹. Furthermore, the “oil, salt and sugar” KAP questionnaire correlated with two more nutritional knowledge questionnaires^24,25 to assess the test–retest reliability by using a correlation coefficient to demonstrate that the results were consistent over time. Our results seemed to be among the best reported in the literature for the KAP questionnaire about the “oil, salt and sugar” in general, varying from 0.50 to 0.90. These results supported the evidence of temporal stability for the KAP questionnaire.

Validity measures whether the comprehensive evaluation system can accurately reflect the purpose and requirements. Refers to the degree to which a measurement tool can measure the correctness of its intended feature. The higher the validity, the better the measurement results can show the characteristics it is intended to measure. In exploratory factor analysis, nine common factors were extracted, with a total of 26 items, and the cumulative contribution rate was 70.9%; a similar value was obtained in the study in India²⁶. This questionnaire was comprehensive and covered the knowledge of oil, salt, sugar, and health and sources of knowledge. Additionally, it covered attitudes toward using these substances in cooking and habits and dietary behaviors to reduce their consumption in daily life. This compensated for the lack of a comprehensive KAP survey on oil, salt, sugar and health in previous studies^27,28,29,30.

This study has several strengths. First, it involved several rounds of group discussion and expert consultation in developing the questionnaire. Second, it had a comprehensive questionnaire composed of several aspects in the knowledge part: the source of knowledge, the relationship between oil or salt or sugar with health, recommended intakes, and identifying unhealthy food with high fat, salt or sugar. The attitude and practice items are based on knowledge items. However, there are limitations related to the study design. Limitations of the study include the small sample size, as many argue that sample sizes of 300 or more are needed for questionnaire development and validation¹⁸. However, the sample size of this study was considered sufficient and fulfilled the criteria to perform the appropriate statistical tests for the questionnaire validation. Additionally, the self-reported questionnaire was the primary source for collecting the information from the canteen staff, and the quality of the data delivered by the study participants was arguable.

For further studies, this questionnaire can be applied to assess KAP related to using oil, salt, and sugar among canteen staff in other areas of China. However, the questionnaire’s applicability among canteen staff outside China requires further examination. As the participants were Chinese, numerous items were developed according to Chinese guidelines and resources. Thus, additional test validations and modifications are necessary to confirm applicability beyond China.

A new validated questionnaire was developed for determining KAP about “oil, salt and sugar” among canteen staff. It was shown to have adequate validity and reliability. Thus, it is a valid questionnaire that assesses knowledge, attitudes and practice levels among canteen staff in China about “oil, salt and sugar”.