Nutrition knowledge, food security, and other risk factors in a sample of college students in Jordan: A cross-sectional design

Abstract

This cross-sectional research assessed nutrition knowledge in a sample of college students in Amman, Jordan and its association with food security and other risk factors. An Arabic Nutrition Knowledge Index (ANKI) was developed and validated in 122 college students (pilot study). In the replication study, the demographics scale, validated ANKI, and Arabic Individual Food Insecurity Experience Scale (FIES) were administered to 470 students. The ANKI was reliable (Cα = 0.82) and stable over time (ICC = 0.82, P < 0.001; paired t-test P = 0.57). Mean ANKI score was good for the pilot study (19.52±0.48) and replication study (19.68 ± 0.22). In the replication study, mean ANKI score was significantly larger among food secure (20.07 ± 0.23) than among moderately (18.28 ± 0.63) and severely food insecure students (14.22 ± 0.70) (F = 11.87, P = 0.000). Thus, measuring nutrition knowledge using the ANKI would provide baseline data for researchers and policy-makers to develop dietary guidelines for Jordanians. Finally, interventions are needed to raise nutrition awareness among college students, to maintain a healthy lifestyle.

Keywords:

• Jordan
• students
• Arabic Nutrition Knowledge Index (ANKI)
• nutrition knowledge
• food security/insecurity

PUBLIC INTEREST STATEMENT

The Arabic Nutrition Knowledge Index (ANKI) is a valid tool to discern nutrition knowledge. The current sample of Jordanian students had good levels of nutrition knowledge. Moreover, nutrition knowledge was better in food-secure students than in the food-insecure. Thus, measuring nutrition knowledge can assist in developing dietary guidelines and nutrition interventions for Jordanians to reduce malnutrition, obesity, and associated chronic diseases.

1. Background

Nutrition knowledge encompasses awareness about the nutritive value of food items and labeling them as healthy or unhealthy. This concept reflects one’s ability to live a healthy lifestyle via improving dietary behaviors (Freeland-Graves & Nitzke, 2013). In line with this, a national study demonstrated that Americans (20–65 years) who had low nutrition knowledge purchased processed foods, consumed large amounts of food items containing saturated fats, salt, and sugar. These individuals were also less likely to meet the recommended dietary guidelines for whole grains, fruits, and vegetables, and thus had low intakes of dietary fiber, as compared to participants with high knowledge levels (Wang & Chen, 2012). Another cross-sectional survey conducted in 1,092 Canadian men and women reported that nutrition knowledge had a positive association with diet quality and adherence to the Canadian dietary guidelines (B = 0.14, p < 0.0001) (Carbonneau et al., 2021). In Jordan, a recent study found that most of the 672 males and females aged 18–34 year were oblivious to the level of wholesomeness or unhealthiness of their diet (Alhaj et al., 2021). Other investigations also evaluated nutritional knowledge in diabetics (El-Qudah, 2016), and athletes (Elsahoryi et al., 2021), as well as about the intakes of salt (Alawwa et al., 2018), supplements (Basheer et al., 2021), and antibiotics (Shehadeh et al., 2012). Yet, none of the previous studies examined nutrition knowledge in Jordanian college students. Moreover, the literature has numerous tools that assess nutrition knowledge (Acheampong & Haldeman, 2013; Barbosa et al., 2016; Klohe-Lehman et al., 2006) or literacy of eating behaviors (Begley et al., 2018; Poelman et al., 2018; Rhea et al., 2020), but none of them are sensitive to the Arabic culture. In addition, there is an Arabic version of the Sports Nutrition Knowledge Questionnaire (Elsahoryi et al., 2021) and General Nutrition Knowledge Questionnaire-Revised (Bataineh & Attlee, 2021; Kliemann et al., 2016; Parmenter & Wardle, 1999). However, they do not contain any Arabic food, and the latter is a long (86 items) time-consuming scale (Bataineh & Attlee, 2021) that may reduce the response rate. Thus, it is important to assess the level of nutrition knowledge of college students in Jordan.

College students are an optimum target group because personal views regarding life choices and routines are formed at this age (Cha et al., 2014; Pelletier et al., 2014; Stran et al., 2016). Consequently, at this phase individuals become more interested in the kinds of foods they eat, maintaining a healthy weight, and following diet trends, especially the ones on social media (Al Ali et al., 2021; Carrotte et al., 2015). Furthermore, these young adults are the future parents whose nutrition information and eating attitudes will affect the dietary behaviors and health status of their offspring (Freeland-Graves & Nitzke, 2013; Kuhl et al., 2012). In Jordan, about 20% of the population (which is ~11-million) is between the ages of 15–24, and the literacy rate is 98%. Over 280,000 students also are enrolled in the 31 Jordanian universities and almost 28,000 are studying abroad (Department of Statistics of Jordan, 2021). The University of Jordan is the primary university in which its students represent less than a quarter of the national number of students (Department of Statistics of Jordan, 2021). In addition, research regarding nutrition knowledge among Jordanians in general and in college students in particular is limited.

On the other hand, food security is the financial capability to buy foods of high nutrient value in amounts that covers the needs of a household (Begley et al., 2019). Purchasing and preparing healthy meals is found to be associated with food security in 1,433 Australian males and females (Begley et al., 2019). Jordan is a low-middle income Arabic country that has undergone social and cultural changes. Such changes mainly affected the citizens of Amman, the capital and the most modernized city in the kingdom (World Food Program of the United Nations, 2019). These developmental modifications altered the lifestyle and eating behaviors of Jordanians, particularly college students. This is probably due to the spread of Western and American retail fast-food restaurants in the area surrounding colleges and universities. For instance, 61% of the adults were overweight and obese (United Nations-Jordan, 2021), and 9.5% lacked food security (Global Food Security Index, 2023). Additionally, 37%, 12%, and 6% of Jordanians died from cardiovascular diseases, cancer, and diabetes mellitus, respectively, in 2018 (Department of Statistics of Jordan, 2021; The World Health Organization, 2018). Therefore, it is essential to assess nutrition knowledge status in college students in Jordan and its relationship with food security.

In all, the primary aim of this study is to document nutrition knowledge in a sample of students from the University of Jordan in Amman using a tool that is sensitive to the Arabic and Jordanian culture, via including traditional food items (healthy and fast foodstuff). The second objective is to measure the association between nutrition knowledge and food security in a sample of Jordanian college students.

We hypothesize that the level of nutrition knowledge would be better in food secure students, and that food security would have a positive association with nutrition knowledge.

2. Methods

2.1. Design and nature of study

Pilot study and replication studies were conducted in random samples recruited from the University of Jordan in Amman, Jordan in the fall and spring of 2021/2022, respectively. Primarily, researchers performed the pilot study to develop and validate the Arabic Nutrition Knowledge Index (ANKI) in a sample of 200 students from the University of Jordan. In the replication study, 470 students (different from the 200 participants of the pilot study completed the validated ANKI from the same institution to (1) document nutrition knowledge levels after re-evaluating the internal reliability of the scale, and (2) discern the association of nutrition knowledge with food security and other risk factors. The protocol of this research was approved by the Institutional Review Board at the University of Jordan, Amman (number: 2021–89), and a written informed consent was obtained from all participants via email.

2.2. Participants

For the pilot study, power analysis showed that a one-tailed t-test required a sample size of 111 students to yield 95% power. However, considering a response rate of 50%, 200 students were contacted. On the other hand, the replication study recruited a random sample of 470 students based on that the students body of the University of Jordan in 2020/2021 was 46,951 (Department of Statistics of Jordan, 2021) and using Taherdoost (2017) sample size equation.

In coordination with the IT department of the University of Jordan, 200 and 470 students were randomly selected from a pool of lists containing student ID numbers. These students, who were from the health, scientific, and humanities faculties, received an invitation email to participate in this research. Inclusion criteria for the focus group, pilot study, and replication study were being a Jordanian, an adult (≥18 years) male and female, an Arabic speaker, and a regularly registered student in the University of Jordan. Hence, individuals were excluded if they were not Jordanian citizens, and were university employee, or student irregularly registered in the university such as attending one specific course. Accordingly, 150 students were eligible for the pilot study, of which 122 adults completed the ANKI twice, with 1-week interval. Meanwhile, the qualified sample for the replication study was 442 students, yet only 414 participated in our research. Due to lack of funding, participants of both studies were not incentivized; nevertheless, those who completed the survey received an educational document containing information about food groups and their nutritional benefits.

2.3. Questionnaire development

The Arabic culture in general and the Jordanian kitchen in particular contain several kinds of nutritious traditional foods like spinach, jews-mallow, mujadara (cooked rice with lentils, onions, and olive oil), labaneh (strained yogurt usually eaten at breakfast), mansaf (lamb meat with cooked rice mixed with Jameed sauce that is made of fermented dried yogurt), shawerma (slices of meat or chicken wrapped in flat bread known as shrak), or falafel (Al-Mughrabi et al., 2019; Bawadi et al., 2012; Girl Eat World, 2021; Mousa, 2019; Tukan et al., 2011). Therefore, an Arabic Nutrition Knowledge Index (ANKI) was developed and administered in a sample of college students. An assistant professor of nutritional sciences generated the items based on a literature review. The items of the ANKI were developed to measure the person’s information regarding food groups, food sources of nutrients (macro- and micro-nutrients), healthy and unhealthy food items (e.g.; whole grains, dietary fibre, fruit juice/drinks, sugar-sweetened beverages, or water), traditional meals and fast foods, American/Western fast foodstuff, coffee intake (Arabic/Turkish vs. American, cappuccino, or instant coffee), and daily physical activity guidelines for substantial health benefits. The generated items of this self-administered scale are in the form of multiple-choice questions/statements, in which every item has four options. Each correct answer has a score of one, where a higher total score reflects greater nutrition knowledge.

Then, a panel of six nutrition professionals (professors in nutritional sciences and dietetics) established their operational definition. T.M. personally asked these professionals, who were from the Department of Nutrition and Food Technology at the University of Jordan, to evaluate the face validity of the ANKI. After obtaining a verbal approval, the professors received a closed envelop of the scale. Moreover, this panel assessed readability, difficulty, content, and bias of the items of the ANKI. Post panel evaluation, the questionnaire included 34 items due to the removal of repeated questions.

Next, a focus group of 15 students was contacted by email, who were chosen randomly from student lists based on their student ID numbers. Only 10 students participated, who met with T.M. in a meeting room in the university to provide their comments about the scale. They evaluated the comprehension and readability of the items of the ANKI via identifying any word or statement that needed clarification. At this phase, the items that did not have strong theoretical support were also excluded. The notes of the focus group and expert panel included: (1) adding the Arabic name of Trans fats, and examples on green leafy vegetables and saturated fats in the questions that its answers mentioned these foods and nutrients; (2) substituting some words with better synonyms such as the word “foods” was replaced with “meals” when asking about eating out; (3) explaining the meaning of “adverse effect” or “health benefit” or “comber potato” in some statements by adding examples (weight gain, weight loss, and baked potato stuffed with cheese, butter and veggies, respectively); (4) rearranging the items according to their type such as grouping the ones that inquire about the sources of nutrients together; and (5) items “purchasing processed foods irrespective of their price”; “stocking empty caloric foods such as cookies, crackers, chips, or chocolate”; and ”engaging in mindless eating like eating a whole pack of chips (7.5 oz) while watching TV” were excluded as they reflected eating behaviors rather than nutrition knowledge. Subsequently, face validity was completed after incorporating the qualitative input of the expert panel and focus group into the final version of the index, comprising 34 statements.

Post the psychometric analyses of the pilot study, the final validated ANKI scale consisted of 30 items, with a total score ranging between 0 and 30. A larger total score indicates better knowledge levels. For nutrition knowledge categorization, the ANKI total score was based on the median (=20). Thus, ANKI < 20 reflects low knowledge levels, whereas an ANKI ≥ 20 indicates high levels. This self-administered scale, which is used in the replication study, takes 10–20 min to complete.

2.4. Pilot study

The refined ANKI was tested in the pilot study. The Director of Infrastructure and Network Department, the Information Centre at the University of Jordan, contacted 200 students by email that was written and signed by T.M. The email asked the students to join this investigation, and a link to a secure research-based website that is accessed only by the students of the University of Jordan. The link directed students to a page that enclosed description about the nature and risks of the study, and a consent form. Students who agreed to participate were directed to the survey page. Only 122 students completed the scale twice with 1-week interval. The obtained data were used to assess the construct validity, internal consistency, and test-retest reliability (degree of consistency over time) of the ANKI.

2.5. Replication study

Similar to the procedure used for the pilot study, every student received the invitation email, asking him/her to participate in the research. The email, which was written and signed by the first author, contained (1) a description about the nature and risks of the study, (2) a consent form, and (3) the survey link to a secure research-based website. The survey collected information from the 414 students about demographics, nutrition knowledge, and food security status. Only the students of the University of Jordan could access the survey link.

2.6. Additional tools of assessment

A constructed self-administered demographic questionnaire obtained information about governorate, age, sex, weight, height, academic year, specialization, grade point average (GPA), employment, income, medical history, physical activity, coffee intake, and smoking.

The Arabic Individual Food Insecurity Experience Scale (FIES) evaluated the food security status of the participants (Ballard et al., 2013; Cafiero et al., 2018). This questionnaire needs 5 min to answer the eight items that asks about the financial ability to purchase food, and availability of food at home. The FIES has a total score range of zero to eight. There is an inverse relationship between the total score and food security status. Food insecurity status is categorized into (1) severely food insecure, (2) moderately food insecure, and (3) food secure (Ballard et al., 2013; Cafiero et al., 2018; Elsahoryi et al., 2020; Tufts University, 2015).

2.7. Statistical analysis

The collected data were entered and analyzed using the Statistical Package for Social Sciences (SPSS 19; IBM, 2010). A thorough univariate analysis was conducted to ensure the validity of the entered data, which were inspected for outliers and possible missing entry. A multivariate diagnostic test (Littlre, 1988) was used to explore the degree of randomness in the identified missing data points, and accordingly the missing data were imputed based on Munro’s methods (Munro, 2005). All tests assumptions firstly were examined before running the analyses. Descriptive statistics were presented as means ± standard error of the mean (SEM) and frequency distributions.

In the pilot study, construct validity was estimated to contrast the nutrition scores on ANKI for nutrition and non-nutrition major students. The degree of internal consistency and test-retest reliability for the ANKI were determined using Cronbach’s α (Cα) value of ≥0.6 and an insignificant paired t-test (P > 0.05), respectively. Furthermore, intra-rater reliability was measured using Cohen’s kappa (Cohen, 1960) in which Kappa values ≤0 indicate no agreement, 0.01–0.20 none to slight agreement, 0.21–0.40 reflect fair agreement, 0.41–0.60 as moderate agreement, 0.61–0.80 reveal substantial agreement, and 0.81–1.00 as almost perfect agreement (Cohen, 1960). Exploratory factor analysis had an acceptable sampling with KMO = 0.711, and significant sphericity (Bartlett’s test of P < 0.001). Factor analysis then was assessed via principal axis factoring with oblimin rotation using 50 iterations and Eigenvalues >1 (Costello & Osborne, 2005). A priori classification of subscales was confirmed, and the number of latent variables and factors of the ANKI was established. Factor loadings of items ≥0.39 were retained in the final version, revealing 10 factors. Nevertheless, three of these dimensions had weak Cronbach’s α values. Thus, we conducted another Factor analysis with oblimin rotation and defining two, three, four, five and six as the number of extracted factors. Except that choosing two, three, four, or five explained less than 60% of the variance. Consequently, the final version of ANKI retained 6 sub-scales and 30 items. Four items [“eating whole grain foods such whole grain bread, fruits and vegetables helps is healthy”; “eating fried foods are not healthy food choices”; “ability to say no to tasty foods”; and “choosing food items according to mood”] were excluded due to having small factor loadings (0.21–0.33).

For the replication study, differences among participants were examined using analysis of variance (ANOVA) for continuous variables. We also applied chi-square test and cross-tabulations to test for differences between categorical variables. Tukey’s and Mann–Whitney U post hoc tests were used to compare two or more independent samples of equal or different sample sizes. Associations of nutrition knowledge with food insecurity and demographic factors (age, BMI, GPA … etc.) were determined using linear, multiple, and logistic regression models. We also evaluated relative risk (RR) for categorical data (such as gender) using 2 × 2 table cross-tabulations, which are presented as odds ratios with 95% confidence interval (CI). Statistical significance was set at P < 0.05.

3. Results

3.1. Pilot study

General characteristics of the students are presented in Table 1. Participants reported a mean age and body mass index (BMI) of 20 years and 22.5 kg/m², respectively. The population sample was from the 12 governorates of Jordan, who had a mean ANKI score of 19.5 with 27 as the highest score.

Table 1. General characteristics of the students of the pilot study (n = 122) and replication study (n = 414)^.*

Variable		Pilot Study (n=122)		Replication Study (n=414)
		Mean (SEM)	Range	Mean (SEM)	Range
Age (years)		19.93 (0.13)	18.0–24.0	20.62 (0.17)	18.0–38.0
BMI1 (kg/m2)		22.49 (0.24)	18.69–29.52	22.84 (0.21)	16.53–43.25
ANKI score2		19.52 (0.48)	6.0–27.0	19.68 (0.22)	6.1–27.3
Governorates		Frequency (%)		Frequency (%)
Central region	Amman	35 (28.7)		125 (30.2)
	Madaba	32 (26.2)		106 (25.6)
	Balqa	24 (19.7)		84 (20.3)
	Zarqa	18 (14.8)		63 (15.2)
Northern region	Irbid	16 (13.1)		57 (13.8)
	Ajloun	14 (11.5)		51 (12.3)
	Jerash	12 (9.8)		45 (10.9)
	Mafraq	11 (9.0)		43 (10.4)
Southern region	Karak	10 (8.2)		36 (8.7)
	Tafilah	9 (7.4)		31 (7.5)
	Ma’an	8 (6.6)		29 (7.0)
	Aqaba	7 (5.7)		27 (6.5)

¹BMI: Body mass index

²ANKI: Arabic Nutrition Knowledge Index

*Data is presented as means (SEM), range, and frequency (%)

The majority of the participants was single (97%), and physically active (75.7%). Over half the students were females (54.9%), and 59.4% of their families had an income less than or equal JD 2,000, and only 6.54% worked (in a food outlet). Participants also indicated that they suffered neither from food allergies nor from any health problem. In addition, the scale was reliable as it had an internal consistency of Cα of 0.82, and perfect agreement reflected by Cohen’s Kappa of 1.0. Further, test-retest of the ANKI at times T1 and T2 had an intra-class correlation coefficient of 0.82 (95% CI [0.78–0.87], P < 0.001) and an external reliability of an insignificant paired t-test (t = −0.576, P = 0.57), indicating stability over time. It is concluded that ANKI is appropriate to measure nutrition knowledge. Moreover, 46% of the participants were nutrition major and 54.1% were studying business, sharia (religious law of Islam), educational sciences, law, languages, international studies, archaeology and tourism, dentistry, science, engineering, or IT. Hence, an assessment of construct validity was conducted, indicating that total ANKI score was significantly higher among nutrition students (23.34 ± 0.71) than these majoring in other specialties (17.97 ± 0.52) (F = 9.38, P < 0.001). The mean difference was 6.34 with an eta squared of about 0.4, reflecting large effect size.

Table 2 illustrates the factor structure of the ANKI administered to the students. The six dimensions had Cα values ranging between 0.61 and 0.82 and explained 65.8% of the variance. The first construct assessed the ability of individuals to identify healthy from unhealthy food choices. The second described participant’s knowledge about local foods rich in essential nutrients and dietary fiber. Nonetheless, the third measured living a healthy lifestyle and eating foods with health benefits. Student’s understanding about the fulfilling healthy food items was evaluated via the fourth dimension. The fifth sub-scale highlighted the intake of meals or foods that maintain wellness, whereas the sixth discerned consumption of healthy meals and snacks. Additionally, five and nine statements are loaded on the first (Healthy Food Choices) and second (Local Nutrient Rich Foods) factors, respectively. Four items also clustered around the third (Lifestyle and Foods with Health Benefits), fourth (Foods and Fullness), fifth (Meals or Foods and Wellness), and sixth (Healthy Meals and Snacks) constructs. Answers of some statements included: “a plate of mujadara with salad” or “a falafel sandwich” or “shawerma sandwich” or “a plate of mansaf with lamb meat” - for item 5. Options of item 6 were “a quarter of vegetarian pizza” or “comber potato” or grilled chicken sandwich’ or ‘beef burger sandwich with fries.” Yet, choices of item 11 consisted of ‘a boiled egg with labaneh’ or ‘a bowel of milk with corn flakes’ or ‘a cup of Nescafé with coffee mate and a cookie’ or ‘a white toast bread with spread butter and jam,’ whereas item 14 contained ‘two dates with a handful of unroasted almonds’ or 2 cups of popcorn” or “a piece of fun size chocolate” or ”2.8-oz bag of chips.” Lastly, items 22 and 23 incorporated “carbohydrate,” or “fat,” or “protein,” or “vitamins and minerals,” as well as “pasta,” “hummus,” “fish,” or “Nabulsi goat/sheep cheese,” respectively.

Table 2. Factor structure of the ANKI administered to the students of the pilot study (n=122)

Item #			Items by Factor		Communalities (h^2)	Pattern Matrix Coefficient	Structure Matrix Coefficient	Cα	Variance (%)
Factor 1 - Healthy Food Choices								0.82	22.24
8			Food item rich in healthy fats		0.73	0.73	0.80
27			Nutrient difference between whole, low-fat, and skimmed milk		0.54	0.74	0.73
19			Food item rich in sodium		0.66	0.68	0.72
15			Difference between whole grain and white bread		0.52	0.62	0.65
2			A healthy food component present on the food label		0.81	0.62	0.60
Factor 2 - Local Nutrient Rich Foods								0.76	13.13
18			Local food item rich in potassium		0.76	0.83	0.82
26			Local food item poor in vitamin B12		0.68	0.76	0.75
24			Local food item rich in vitamin A		0.55	0.64	0.62
29			Local food item or meal rich in dietary fiber		0.44	0.58	0.61
25			Local food item poor in iron		0.62	0.60	0.61
20			Local food item rich in calcium and vitamin D		0.31	0.55	0.57
23			Local food or meal rich in protein		0.75	0.43	0.48
17			Local food item rich in vitamin C		0.83	0.69	0.42
21			Local food item or meal rich in zinc		0.83	0.39	0.42
Factor 3 - Lifestyle and Foods with Health Benefits								0.70	9.68
7			Minimum daily physical activity to stay fit		0.64	0.78	0.71
28			Bread is rich in which nutrient		0.73	0.66	0.67
3			Number of water cups must be consumed per day		0.48	0.41	0.45
10			Coffee type with the least adverse health effects		0.70	0.43	0.42
Factor 4 - Foods and Fullness								0.66	7.46
11			Most satiating meal for breakfast		0.58	0.68	0.67
22			Best nutrient to cover our daily energy needs		0.60	0.54	0.52
14			A nutrient-dense meal/snack		0.41	0.44	0.46
16			Proportion of fruits and vegetables in a meal or plate		0.65	0.47	0.40
Factor 5 - Meals or Foods and Wellness								0.63	7.14
1			Daily number of meals to maintain a healthy weight		0.65	0.59	0.62
12			A sweet item or dessert with the lowest caloric content and least adverse health effects		0.52	0.50	0.52
30			Fish is a good source of which kind of fat		0.76	0.49	0.44
13			Sugar content in a 340 ml can of carbonated beverages		0.43	0.47	0.42
Factor 6 - Healthy Meals and Snacks								0.61	6.11
6			A traditional fast food with the lowest fat content		0.73	0.83	0.80
5			A traditional meal with the lowest fat content		0.70	0.75	0.76
9			Unhealthy fruit drink or juice		0.70	0.58	0.40
4			A healthy soft drink		0.53	0.43	0.40

*ANKI: Arabic Nutrition Knowledge Index

3.2. Replication study

The students had similar mean age, BMI, and ANKI score to that of the pilot study (Table 1). Other characteristics of this sample were also comparable to these of the pilot study. For example, 92% were single, 69% exercise, and about two-third were women. Further, 67.6% of their families had an income ≤ JD 2,000, 11.6% of the students were employed, and none suffered from any health problem or food allergy. Moreover, the internal consistency of ANKI was in agreement with that of the pilot study as reflected by Cα value (=0.85), suggesting high internal reliability.

Approximately, 85.5% of the participants scored ≥15, and 56.3% obtained a total score of ≥20 on the ANKI. Figure 1(1) shows the percentage of the students answering correctly on the items of the ANKI. Six items [items 1 (meal frequency), 3 (water intake), 6 (traditional fast foods), 9 (unhealthy fruit drink or juice), 15 (whole grain vs. white bread), and 16 (fruits and vegetables)] had lower proportion of correct responses (34%–42%); whereas more than half of the students (51%–77%) answered 24 questions right. For example, 74.6% of the population sample selected the black Arabic, Turkish, or American coffee as the healthiest coffee type in comparison to drinking sweetened coffee, 2in1 Nescafé, or cappuccino.

Figure 1. The percentage of students of the replication study answering correctly on the items of the ANKI (n=414).

Figure 2 illustrates the relationship between nutrition knowledge level, represented by ANKI, and food security status and demographic variables. More than half of the participants who had greater nutrition knowledge levels aged 18–20 years and had a GPA > 3 (P < 0.05). Better nutrition knowledge status were also seen in about one-third of the students who were in their first academic year, had a family income >2000 JD, and drank Arabic/Turkish coffee or Americano (P < 0.05). Approximately, 82.8%, 74.3%, 66.1%, 42.5%, and 5.6% of the population sample with high nutrition knowledge significantly were non-smokers, exercised, had normal weight, ingested supplements, and suffered from allergy, respectively (P < 0.05). In relation to food security, 85% of the students with larger levels of nutrition knowledge were food-secure, and none of these participants lacked food security (X² = 54.9, P < 0.05). In relation to sex, there was an insignificant difference between the number of females (64%) and males (36%) who had high and low nutrition levels (X² = 1.13, P = 0.17) (data is not shown).

Figure 2. The relationship between nutrition knowledge level (represented by ANKI) and food security status and demographic variables for the replication study (n=414).

^{a,b,c,d,e,f,g} Different superscripts indicate significant differences within categories of demographic variables and food security status for students with high nutrition knowledge levels versus those with low knowledge levels.

Table 3 also shows that nutrition knowledge score was significantly greater in students who were in their first, third, and fourth academic years, and whose major was nutrition as compared to those in their second year and who were specialized in non-nutrition studies (P < 0.05). Mean ANKI score also was significantly larger among food secure than among moderately and severely food insecure students (F = 11.87, P = 0.000).

Table 3. Nutrition knowledge (ANKI) mean scores for the replication study (n=414)^.*

Variable	Mean (SEM)	F	P-value
Sex		0.675	0.412
Women	19.82 (0.29) ^a
Men	19.44 (0.29) ^a
Year of Study		23.59	0.000
1st	19.43 (0.35) ^a
2nd	17.78 (0.39) ^b
3rd	19.98 (0.51) ^a
4th	22.80 (0.36) ^c
Family monthly Income (JD)		7.86	0.000
≤400–2000	20.02 (0.28) ^a
2000 - ≥ 4000	21.03 (0.64) ^a
do not wish to answer	18.25 (0.37) ^b
Specialization		9.38	0.000
Nutrition	23.34 (0.71)^a
Other specializations	17.97 (0.52)^b
Weight status (BMI in kg/m2)		1.601	0.189
Underweight (BMI= below 18.5)	20.46 (0.99) ^a
Normal weight (BMI = 18.5–24.9)	19.40 (0.28) ^a
Overweight (BMI = 25–29.9)	19.89 (0.36) ^a
Obese (BMI = ≥ 30)	21.50 (0.39) ^a
Physical activity		0.060	0.807
No activity	19.60 (0.41) ^a
Active	19.72 (0.26) ^a
Food security status		11.87	0.000
Food secure	20.07 (0.23) ^a
Moderately food insecure	18.28 (0.63) ^b
Severely food insecure	14.22 (0.70) ^c

*In columns, different superscripts indicate significant differences within categories

Regression models in Table 4 present the association of nutrition knowledge with food security status (reflected by FIES score) and other variables for the population sample. Nutrition knowledge was positively associated with BMI, GPA, and the daily number of coffee cups consumed (P < 0.01). Yet, nutrition knowledge was negatively associated with the daily number of cigarettes, and FIES score (P < 0.05). Furthermore, the inverse association between nutrition knowledge and FIES score reflects a positive association between food security status and nutrition knowledge levels, since the higher FIES score the greater food insecurity (P < 0.05). Table 4 also demonstrates that students who were older and consumed more cups of black coffee per day (after controlling for all variables) were more likely to have better nutrition information when compared with those who had low knowledge levels (P < 0.01). In addition, for every unit decrease in BMI and GPA, the odds of acquiring good nutrition knowledge have increased by 10.7% and 16.9%, respectively (P < 0.01).

Table 4. Association between nutrition knowledge (using ANKI) and food security status (reflected by FIES score) and other variables for the population sample of the replication study (n=414).*

Indicator	β^‡	P	OR (95% CI)	P
Age (years)	0.119	0.053	1.135 (1.050–1.227)	0.001
Sex**	__	__	0.802 (0.534–1.205)	0.303
BMI (Kg/m^2)^‡	0.779	0.000	0.893 (0.823–0.969)	0.007
GPA^†	1.841	0.000	0.831 (0.767–0.943)	0.000
Number of cigarettes/day	−0.212	0.000	1.051 (0.979–1.129)	0.166
Number of coffee cups/day^¶	0.663	0.002	1.451 (1.178–1.787)	0.000
Food security (FIES score)^§	−1.094	0.032	0.967 (0.863–1.085)	0.571

*Data is considered statistically significant at P< 0.05. The OR increases for every unit increase in the predictor (independent variable). Reference group is low nutrition knowledge (ANKI < 20) (OR =1).

ANKI: Arabic Nutrition Knowledge Index

**Men as the reference group.

^‡BMI = body mass index

^†GPA = grade point average

^¶Controlled for all variables.

^§FIES: Individual Food Insecurity Experience Scale. A higher score reflects greater food insecurity.

4. Discussion

4.1. Pilot study

The present study showed that ANKI is a valid and reliable scale to measure nutrition knowledge in an Arabic-speaking population as reflected by its good internal and external reliability. For instance, its internal consistency (Cα of 0.82 and 0.85 for the pilot study and the replication study, respectively), intra-class correlation coefficient (ICC) (0.82, P < 0.001), and external reliability (P > 0.50) were high. These results are in accordance with these of the Arabic version of the General Nutrition Knowledge Questionnaire (Cα = 0.91, ICC = 0.84, and P > 0.3) (Bataineh & Attlee, 2021), despite the difference in the items content in which the current ANKI is more sensitive to the Arabic/Jordanian cultural foods.

Moreover, students specialized in nutrition had significantly better nutrition knowledge levels than those who lacked such background (P < 0.001). This indicates that the ANKI is a well-constructed scale. Our findings are in parallel with previous studies that assessed nutrition knowledge using General Nutrition Knowledge Questionnaire-Revised in college students in Jordan, the United Arab Emirates, Kuwait (Alkaed et al., 2018), Turkey (Al Saffar, 2012), Uganda (Bukenya et al., 2017), and England (Kliemann et al., 2016; Parmenter & Wardle, 1999).

4.2. Replication study

It is worth noting that Jordan lacks a country profile dietary guidelines (Food and Agriculture Organization of the UN FAO. Food-based Dietary Guidelines, 2022; UN-Jordan, 2021), and therefore it follows the dietary guidelines of the World Health Organization (The World Health Organization WHO, 2020). In the present study, about two-third of the participants chose three meals as the best frequency for a healthy lifestyle (Al-Mughrabi et al., 2019; Bawadi et al., 2012; Girl Eat World, 2021; Mousa, 2019; Tukan et al., 2011). This is not in agreement with findings reported in the literature that having small frequent meals such as three meals and 2–3 snacks/day is a better regimen to maintain wellness (Al-Domi et al., 2021; Schoenfeld et al., 2015; WHO, 2020). Similarly, 65% believed that drinking 2–7 cups of water on daily basis is enough. This is, however, less than the recommended for an adult, an average of 8 cups/day, to achieve optimum health (The World Health Organization WHO, 2017). These results are inconsistent with the findings of Al-Domi et al. (2021) in which 69% of 4,388 adults consumed over 8 cups per day.

In the Jordanian market, there are three types of fruit juices including sugar-sweetened drink (≤10% fruit), processed juice or fruits’ nectar juice (~60% fruit), and fruit juice with no added sugar (100% fruit) (Jordan Food and Drug Administration, 2022). Participants were unaware about such differences, mainly that the majority selected the processed fruit juice (second type) as the most drink with adverse health effects. In addition, white bread and rice are Jordanian staple foods that account for at least half of the plate; this would explain the low consumption of whole-grain bread. Moreover, fruits and vegetables form one-quarter to one-third of the Jordanian plate/meal (Al-Mughrabi et al., 2019; Bawadi et al., 2012; Mousa, 2019; Tukan et al., 2011). Most students of this study indicated following this practice. Such behaviors, however, do not conform to the recommended dietary guidelines (WHO, 2020). We also found that participants mainly chose the black Arabic, Turkish, or American coffee to be the healthiest type. This is because generally, black coffee does not contain sugar, milk, cream, or any other additive that raises its caloric content. In contrast, non-brewed blended coffee have poorer health benefits due to their high caloric content (from additives) (Huang et al., 2009) and lower concentration of phenols (Niseteo et al., 2012), when compared with brewed coffee. The Arabic coffee is prepared by boiling water followed by adding coffee powder, and then boiling it for another few minutes, in which coffee beans are roasted with cardamom prior to grinding. During the boiling process, sugar can be added if the coffee is preferred sweetened. Thus, measuring nutrition knowledge using the ANKI would provide baseline data for investigators and policy-makers to develop dietary guidelines for Jordanians.

Most of our sample had good levels of nutrition knowledge (mean ANKI score ≥15), which is a very important life aspect to achieve optimal health via following a healthy routine. For instance, nutrition mindfulness helps individuals to make healthy food choices that reduce their risk to gain weight and develop chronic diseases. In line with this, a recent research examined the health behaviors of 4,388 Jordanian males and females during the COVID-19 lock-down (Al-Domi et al., 2021). The authors indicated that 77.5% of the participants were overweight or obese. Furthermore, 60.1% consumed over three snacks a day, and 44.3% experienced increased appetite. Data also revealed that 34.2% and 38.7% of the population sample gained weight and engaged in sedentary behaviors, respectively (Al-Domi et al., 2021). Thus, conducting educational interventions is essential to raise nutrition awareness among Jordanians, particularly college students. Consequently, individuals would be able to read food labels, as well as purchase and eat foods of high nutritive value (i.e.; rich in essential nutrients and low in energy, saturated fats, sugars, and salt) (Appelhans et al., 2014; Barbosa et al., 2016; Fitzgerald et al., 2008; Miller & Cassady, 2015; Quaidoo et al., 2018; Wardle et al., 2000).

Furthermore, the present findings revealed that the level of nutrition knowledge was better among those with higher academic level and having good GPA, which is supported by previous research (Meyer et al., 2021; Weerasekara et al., 2020). We also found that the better the nutrition information acquired, the healthier the weight a person has (P < 0.01). This is comparable to the findings of Weerasekara et al. (2020). The authors observed that the level of nutritional knowledge was significantly greater in normal weight 400 Sri Lankan women (P < 0.05) (Weerasekara et al., 2020). Thus, nutritional interventions must be conducted to college students to raise their awareness level about the healthy and unhealthy foods, and the importance of consuming a healthy balanced diet.

Nutrition knowledge was also larger among the current students who consumed coffee, particularly drinking black coffee (i.e.; Arabic/Turkish coffee or Americano) [OR (95% CI): 1.451 (1.178–1.787), P = 0.000]. This is probably because coffee is known to have some health benefits due to containing antioxidants (Niseteo et al., 2012). In addition to that, in Jordan coffee is the most drink served and consumed; especially, that it is associated with social events such as weddings and funerals (Bawadi et al., 2012; Tukan et al., 2011). Moreover, students who engaged in walking had better nutrition awareness status than inactive ones (P < 0.001). This is in agreement with the outcomes of a recent survey that examined nutritional perceptions of 672 Jordanian males and females aged 18–34 years towards eating and physical activity (Alhaj et al., 2021). The levels of nutrition knowledge were also greater in non-smokers than in smokers (P < 0.001). Nonetheless, it is necessity to conduct educational programs about smoking targeting college students in Jordan. This is vital since smoking is common in the Jordanian youth (45–62%), who primarily smoked cigarettes and/or nargilah (Alkouri et al., 0000). This phenomenon also contributes to almost 80% of deaths from non-communicable diseases in Jordan (The World Health Organization WHO, 2021).

This study also indicated that the level of nutrition knowledge was significantly better in food-secure students than in the food-insecure (P < 0.05). Furthermore, food security status was associated positively with nutrition knowledge (P < 0.05), however food security was not a predictor of nutrition knowledge (P = 0.571). Nevertheless, having good nutrition information is essential in guiding individuals to make healthy food choices. Accordingly, people would be able to maintain their health, and reduce the risk of malnutrition or chronic diseases (Barbosa et al., 2016; Quaidoo et al., 2018; Wang & Chen, 2012). Therefore, nutrition interventions such as workshops/sessions targeting college students should be developed. Such programs must provide information about food groups; healthy eating and snacking at home and on the go; food shopping practices; and food management. Future surveys should also assess the nutrition and health status of food secure and insecure Jordanians, mainly of university students, post the pandemic.

The current study had some limitations though including the absence of an Arabic tool to compare with it the findings of the ANKI has limited our capability to measure concurrent validity. Scarcity of funding also prevented incentivizing participants, which limited the size of the population sample, and recruiting students from other universities in Jordan. Yet, conducting a pilot study using a population sample of at least 100 individuals is reported to be adequate (Anthoine et al., 2014; Browen, 1995; Connelly et al., 2008), and repeating it in a larger sample (n = 414) has confirmed the findings of the pilot study. The present data also cannot be generalized due to the cross-sectional design. Accordingly, prospective studies should be conducted to explore the generalizability of the ANKI, or a modified version of it, to other age groups and Arab nations. On the other hand, the strengths of this research are the validation of the first scale that measured nutrition knowledge in a diverse Jordanian sample of university students, which was reliable, stable over time, short, and timely, therefore allow having a high participation rate. In addition to that, this research included 414 participants of both genders from different academic specializations, economic levels, and cities.

5. Conclusions

The ANKI is a valid tool to assess nutrition knowledge in an Arabic-speaking population, particularly Jordanians. Thus, measuring nutrition awareness using the ANKI would provide baseline data for researchers and policy-makers to develop dietary recommendations and guidelines for Jordanians, particularly due to the absence of such strategy. Further, the current sample of Jordanian college students had good nutrition knowledge levels. The findings of this research also showed that nutrition knowledge was better in food-secure students than in the food-insecure. Moreover, the degree of nutrition awareness was positively associated with age, GPA, daily coffee consumption, and food security. Yet, nutrition knowledge was negatively associated with weight status and smoking. Thus, interventional programs are required to spread nutrition awareness among college students. This is because at this age opinions regarding eating habits and lifestyle behaviors, such as smoking and exercising, are formed. Hence, understanding the nutritive value of foods would guide young adults to make better food choices, and prepare healthy meals for themselves and their families. This is crucial to maintain a healthy routine, achieve optimum health, and reduce the risk of food insecurity, malnutrition or obesity and associated chronic diseases.

Availability of data and materials

Zenodo: Data of Nutrition Knowledge in a Sample of College Students in Jordan

https://zenodo.org/record/7698725 (Mousa & Dardas, 2023)

DOI: 10.5281/zenodo.7698725

This project contains the following underlying data:

• ANKI_Research_Dataset.csv

• ANKI_Data.xlsx

And extended data:

• Scale of ANKI

• Studies consent form

• Studies nature

Data are available under the terms of the Creative Commons Attribution 4.0 International license (CC-BY 4.0).

Contributors and authors’ roles

Tamar Y. Mousa developed the design of the research, and the Arabic Nutrition Knowledge Index (ANKI) after reviewing the literature. She also contacted and communicated with the panel of experts, as well as with the focus group. Accordingly, she incorporated the qualitative inputs of the panel groups and focus groups into the final version of the ANKI. In addition, she communicated with the IT department to send the surveys to the students. Finally, she conducted statistical analysis and wrote the manuscript.

Latefa A. Dardas conducted psychometric and statistical analyses and contributed to manuscript writing.

Financial support

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Reporting guidelines

This article followed the STROBE checklist.

Acknowledgments

Authors’ AgreementThe authors declare that this manuscript is original, has not been published before, and is not currently being considered for publication elsewhere. The authors confirm that there are no known conflicts of interest associated with this publication, which did not receive any financial support. The lead author affirms that this manuscript is an honest, accurate, and transparent account of the study being reported. The reporting of this work is compliant with The Code of Ethics of the World Medical Association (Declaration of Helsinki). The lead author affirms that no important aspects of the study have been omitted and that any discrepancies from the study as planned have been explained. In addition, the protocol of this research is approved by the Institutional Review Board at the University of Jordan, Amman, Jordan (ref no.: 2021-89). An informed consent was obtained from all subjects involved in both phases of the study. In addition, permission to use the tools of assessment was obtained from all researchers.

The submitted manuscript has been read and approved by all authors. The authors confirm that there are no other persons, who satisfied the criteria for authorship, but are not listed. The order of authors listed in the manuscript has been approved by all of them. The authors confirm that a due consideration has been given to the protection of intellectual property associated with this work and that there are no impediments to publication, including the timing of publication, with respect to intellectual property. The authors understand that the Corresponding Author is the sole contact for the Editorial process (including Editorial Manager and direct communications with the office), and holds the responsibility for communicating with the other authors about progress, submissions of revisions and final approval of proofs.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Notes on contributors

Tamara Y. Mousa

TamaraY Mousa is an assistant professor of nutrition at the University of Jordan, and a member of the American Society of Nutrition, Supreme National Committee for Nutrition at Ministry of Health, Jordanian Dietetic Association, and Jordanian Food and Nutrition Association; editor-in-chief of Frontiers in Sustainable Food Systems; and reviewer for several indexed journals. I obtained my PhD from the University of Texas at Austin and have over 30 publications on eating disorders, body dissatisfaction, nutrition status, and food insecurity.

Latefa A. Dardas

Latefa A. Dardas is an associate professor of community health nursing at the University of Jordan. I have >50 publications about psychiatry; mental health; bullying; and knowledge, attitudes, and practices toward COVID-19. I got my PhD from Duke University, and I am a member of the Psychological Sciences Association; and a credentialed Psychiatric and Mental Health Nurse with specialization in Child and Adolescent Mental Health (North Carolina, USA), and certified in Advanced Applied Statistics for Health Sciences.