Exploration of Malaysian school-children’s food preferences: what do we know?

Abstract

Background

Feeding practices exert a definite influence over children’s experiences. This article aims to explore parental feeding practices and investigate the prediction domain of food preference from parent–child perspectives.

Methods

Two individual studies were conducted on Malay families with children aged 7–12 years. In Study 1, mothers (n = 17) participated in semi-structured focus group interviews on their knowledge of foods and feeding practices. In Study 2, parent–child pairs (n = 14) answered a 36-item, 5-point Likert scale Food Preference Questionnaire followed by virtual structured qualitative interviews. The interviews were digitally recorded, transcribed verbatim, back-translated and analysed according to the framework analysis technique.

Results

In Study 1, mothers perceived vegetables, chicken, fish and plain water as healthy foods and drinks while discretionary options were snacks, fast foods and carbonated drinks. The mothers defined healthy foods as foods handled safely with health benefits. They used ‘healthy’ cooking methods to prepare preferred foods and overtly controlled the child’s access to discretionary food. In Study 2, the food groups reported by parent–child pair’s report were consistent for the most preferred foods [snacks, median (interquartile range), parent: 4.5 (1.0) vs. child: 4.5 (0.0), p > 0.05] and least preferred food [legumes, parent: 2.0 (1.0) vs. child: 2.0 (1.0), p > 0.05]. Parents emphasized taste as the key determinant of food preference.

Conclusion

These studies were the first to qualitatively explore parents’ perceptions of foods affecting their feeding practices among the Malaysian community to highlight the cultural contribution. Key insights into children’s food intake and factors influencing their food preferences were identified.

food and diet, parenting, children, family, Malaysian

INTRODUCTION

High rates of childhood malnutrition are affecting both low- and middle-income nations in Asia [1], including Malaysia [2]. Concurrently with malnutrition, the rise in the national prevalence of stunting, overweight and obesity amongst Malaysian children is also becoming of great concern [3]. Data from the National Health and Morbidity survey from 2011 to 2019 showed that for children under 5 years of age, stunting has increased by 5% (from 16.7% to 21.8%), and obesity amongst children aged between 5 and 17 years more than doubled (from 6.1% to 14.8%) [3]. Current evidence posits that both undernutrition and overweight can propagate severe long-term health effects, translating into enormous economic implications [4].

Parents and carers were identified to be one of the key predictors of children’s dietary habits [5, 6]. Predominantly in younger children, feeding practices exert a definite influence over their environments and experiences [7], consequently impacting the children’s attitude and behaviour towards their food intake. Feeding practices are proposed to be divided into two broad categories: directive and non-directive [8, 9]. Directive control such as restriction and pressure-to-eat has been associated with unhealthy eating behaviours as they interfere with the child’s innate ability to regulate eating [8, 10, 11]. In contrast, the non-directive control helps promote healthy behaviours and establish a positive relationship with food [8, 10, 11] through positive role-modelling and a healthy food environment [12, 13]. Taken together, the non-directive control in parental feeding practices could be considered as a more supportive approach to foster healthier eating habits in children [14].

While caregivers play a significant role in shaping dietary habits, food preference was found to equally impact the children’s diet quality [15]. Food preference specifically on taste appears to be partially determined by genetics in children, which can contribute to ‘pickiness’ in eating [16]. Pickiness is defined as a refusal to eat specific food groups or try new foods [17] and is more prevalent amongst younger children [18]. Furthermore, there is an innate preference for sweet and salty [19, 20], and avoidance of bitter and sour flavours [21, 22]. Amongst different age groups, the preference and intake of sweet foods were higher in children below 10 years of age, which declined in adolescence and adulthood [23–25]. This can potentially suggest that such innate taste preferences are modifiable by dietary exposure and experience in later life [22], largely due to the impact of the environmental exposures including culture [26]. In Malaysia, the Malay adult population reported a greater preference for a Western-based food pattern that is high in salt, sugar and fat [27, 28].

Previous studies examining feeding practices [29–31] have primarily used specific questionnaires [32, 33] developed from the Western population [34–37]. However, as feeding practices are strongly influenced by environment and culture; children-focused studies remain relatively scarce in the Malaysian context [38]. It is widely accepted that the use of a qualitative approach allows for in-depth exploration to identify the gaps which are essential when seeking to gain insights into picky eating. Therefore, we performed two separate studies exploring parents’ knowledge about food, its effect on their feeding practices and their children’s food preferences.

METHODS

The project design was based on the conduction of two individual and related studies. Study 1 was a qualitative study on parental knowledge of foods, drinks and feeding practices. Study 2 utilized the mixed-method approach to assess additional variables on food preferences from the perspectives of parents and their children using a questionnaire and followed by structured interviews.

In Study 1, participants were recruited using the purposive sampling method from primary schools and local communities in Klang Valley, Malaysia. Ethical approval was obtained from the [removed for blind peer review]. Permission to recruit from local educational institutes and residential communities was obtained respectively where invitation pamphlets were sent to all parents. Eligible participants must be the main caregivers of a child aged between 7 and 12 years of Malay ethnicity. The specific age was selected due to the lack of validated questionnaires investigating child-reported food preferences in an age-appropriate manner and current literature suggests that children aged 6 years and above can accurately report their food preferences [39]. The main caregiver was defined as an adult staying together with the child most days of the week and the primary decision-maker for food practices (purchasing of groceries, food preparation and/or serving) [40]. Consenting caregivers were invited to attend face-to-face semi-structured focus group discussions, conducted between February and March 2020. The interview guide (Table 1) was developed based on the restriction domain of the Child Feeding Questionnaire [33] and an in-depth literature review on feeding practices. The face validity of the interview guide was tested with three participants, conducted by one moderator and an observer. Minor amendments were made to the interview guide based on feedback on the clarity of questions used during the interview. Subsequent interview sessions were performed by a moderator. The focus group consisted of two to five participants at the time. All focus groups were conducted in a private discussion room in the school or community hall. At the start of each focus group, participants were briefed on the study objectives, data privacy and confidentiality. Participants were also asked to complete a brief demographic questionnaire and household income based on the national classifications [41]. Participants were encouraged to express their thoughts freely based on open-ended questions. Each group session was on average 60 min in duration and all sessions were audio and video recorded. Sampling, recruitment and focus group interviews continued until data saturation is reached. All recordings in the Malay language were translated to English and transcribed and checked for accuracy of the translation. The transcribed data were coded into the frequency of responses and organized into several themes and sub-themes. The sub-themes were grouped through the recurring explanations by the parents which represented the thematic units [42]. This thematic analysis process was done manually using the Microsoft Excel spreadsheet by the research team. Final thematic analysis, categorization and conclusions were reached by consensus of all members. Exemplar key quotations were identified within each theme to illustrate the views of the participants.

Table 1.

Discussion questions asked during focus group discussion (Study 1)

Interview guide
1. How do you define healthy and unhealthy foods?
2. What are the foods you think are healthy and unhealthy?
3. When you think the food is healthy, do you allow your child to have it?
a. If the answer is no:
i. Why do you disallow?
ii. How do you disallow?
b. If the answer is sometimes:
i. What makes you decide to give at times?
ii. How do you decide to give?
iii. How often do you allow?
c. If the answer is yes:
i. What makes you decide to allow the children to have it?
ii. How do you decide to give?
iii. How often do you allow it?
4. When you think the food is unhealthy, do you allow your child to still have it?
a. If the answer is no:
i. How did you disallow it?
b. If the answer is sometimes:
i. How did you disallow it?
ii. How often do you still allow?
iii. What makes you decide to still allow it at times?
c. If the answer is yes:
i. How often do you allow your child to have it?
ii. What are the reasons you allow your child to have unhealthy foods?

In the second study which took place during the SARS-CoV-2 pandemic, a semi-qualitative study was conducted virtually for parent–child pairs between February and June 2021. Ethical approval was obtained [removed for blind peer review] and digital recruitment messages were randomly disseminated to recruit free-living, Malay children aged 7 to 12 years old and their parents from Klang Valley, Malaysia. The inclusion criteria were determined as (i) children aged 7 to 12 years old, (ii) in good health condition and (iii) having their parents as their main caregiver (defined as staying together most days of the week and the main decision-maker for food practices). Eligible participants were contacted to provide the child’s assent and parental consent before completing an online questionnaire and the structured interviews. In contrast to Study 1, a questionnaire on food preference was used for more accurate reporting by children and parents to determine children’s food preferences. The list of food and drinks in Food Preference Questionnaire (FPQ) was retrieved from a local dietary study on pre-adolescent Malay children. The 36-item FPQ of the 5-point Likert scale questionnaire (‘dislike a lot’, ‘dislike’, ‘neutral’, ‘like’ and ‘like a lot’) was modified from Jani et al. [40] to meet the requirements of the Malaysian population. During the online interviews, children and their parents were asked individually in detail about the factors affecting the children’s preferences. The questions used were adapted from the Food Choice Questionnaire [43], which was validated amongst Malaysian adolescents [44]. Each interview session was recorded using Microsoft Teams Version 4 (Microsoft Corporation, Kuala Lumpur, Malaysia), consisting of one facilitator and one recorder. Statistical analyses were performed using the SPSS software (v26.0, Armonk, NY: IBM Corp) on the quantitative data (sociodemographic and food preference). For each item in the FPQ, median and interquartile range (IQR) values, number and percentages were calculated. The food preference scores based on the 5-point Likert scale of the parent–child pairs were compared using the Mann–Whitney U test (significance level of p < 0.05). The recorded interviews were transcribed and back-translated. Themes and coding were generated via thematic analysis using Microsoft Excel Version 26 (Microsoft Corporation, Kuala Lumpur, Malaysia) until data saturation was reached [45, 46].

RESULTS

Table 2 summarizes the participant’s sociodemographic background. In Study 1, seventeen participants completed their focus groups in seven interview sessions. All participants were mothers with half of them being housewives and the remaining working (47.1%). About 82.4% of the participants were the main person deciding on the household food practices while others shared the responsibility with their spouses. In Study 2, 14 parent–child pairs participated with more girls (71.4%) than boys (28.6%), and more mothers (85.7%) than fathers (14.3%). Contrary to Study 1, most parents were working full time (57.1%) with a college or university education level (85.7%) with a higher total family monthly income between RM 4500 to RM 10 000 (42.9%).

Table 2.

Characteristics of participants in Study 1 and Study 2

	Study 1 (N = 17)	Study 2 (N = 14)
	Parents, n (%)	Parents, n (%)	Children, n (%)
Gender
Male	NA	2 (14.3)	4 (28.6)
Female	17 (100)	12 (85.7)	10 (71.4)
Age (years)	40.1 (5.0)	42.9 (6.0)	10.0 (4.0)
Employment status
Full time	6 (35.3)	8 (57.1)	NA
Part-time	NA	1 (7.1)
Self-employed	2 (11.8)	4 (28.6)
Others (Housewife)	9 (52.9)	1 (7.1)
Highest level of education			NA
Primary	2 (11.8)	NA
Secondary	11 (64.7)	2 (14.3)
College/University	4 (23.5)	12 (85.7)
Family size (members)			NA
3–4	3 (17.6)	6 (42.8)
5–6	10 (58.8)	5 (35.7)
7 and above	4 (23.5)	3 (21.4)
Total family monthly incomea			NA
<RM4500 (B40)	13 (76.5)	4 (28.6)
RM 4500–RM 10 000 (M40)	2 (11.8)	6 (42.9)
>RM 10 000 (T20)	2 (11.8)	4 (28.6)
Household main food decision-making			NA
Yes	14 (82.4)
Share with spouse	3 (17.6)

NA, not applicable; IQR, interquartile range.

a

Based on the household income group categorization by the Department of Statistics, Malaysia (DOSM) [43].

In Study 1, while all participants reported vegetables were healthy, most participants (10/17) stated fish as a healthier option as compared to prawns and squid. For the definition of healthy foods, most participants (10/17) identified them as foods with beneficial effects and unhealthy foods being harmful to the body. The other elements of healthy foods include being rich in vitamins, protein and fibre, less oily and handled safely. Plain water was mostly reported as a healthy beverage because it was considered ‘safe’ for consumption (8/17) and did not contain any additives in contrast to carbonated drinks (11/17) and sweetened beverages (6/17) named as unhealthy beverages. Common local snacks such as banana fritters, potato chips or fish chips ‘keropok lekor’ (13/17) were most frequently reported as unhealthy food. Eight participants classified fast food as unhealthy mainly attributed to its characteristics of high salt, high fat and high sugar. Food stalls along the roadside and poor food handling were cited as the source of unhealthy food (7/17). Nearly all participants reported they provided healthy food to their children daily. When it comes to unhealthy food, most parents did not restrict their child’s intake and allowed it occasionally (15/17), only two participants completely restricted their child from consuming unhealthy food.

Three key themes (Table 3) were identified from Study 1 on participants’ attitudes towards unhealthy foods; factors affecting child’s food preference, reasons for providing unhealthy foods and avoidance of unhealthy foods. Data saturation was observed when reached the seventh group, indicating that a comprehensive of the topic was achieved. Key theme 1 (Factors affecting child’s food preference): The participants identified various factors affecting the child’s food preference including the child’s liking, cooking methods of the foods, and how foods are served. Most participants stated that their children accepted most, or all foods provided to them. This allowed them to ensure a balanced food intake by providing different types of foods. The participants reported that their children would eat those foods that were prepared to meet the child’s acceptance or preference. Therefore, they would mix perceived healthy foods together with the child’s preferred foods into a single dish to encourage their child to eat healthily.

Table 3.

Themes on parental perception and feeding practices on healthy and unhealthy foods (Study 1)

Key theme	Sub-themes	Responses
1. Factors affecting child’s food preference	Age when exposed to new foods Child own likings Cooking methods of the foods The knowledge provided to the children Eating practices of peers Serving methods of the foods Behaviors of different role models Convince the child to try new foods choices introduced Choose foods with a longer shelf life to make sure the children always get to have those healthy foods	‘I will ask my children what kind of foods they want to eat and how frequently they want to eat because my children eat almost everything. I won’t force them to eat foods they don’t like, but I will make sure they have a balanced nutrient intake while preparing their foods.’ (SKB2-011) ‘Cooking methods of the meal affects the intake of my child’s meal. He eats more fruits than vegetables, therefore I will prepare the foods with the cooking method acceptable to him to make him eat more vegetables. For example, I will cut the vegetables to smaller pieces so that he will eat those vegetables.’ (SKB2-003)
2. Reasons for providing unhealthy foods	Feel sorry for the children, wanted to make the child happy To let the child have a taste of those foods To satisfy the child’s craving and preference Unable to control the child when not under the direct supervision	‘I will still give those foods to my children once a week during the weekends, as the children have been eating all those healthy foods during the weekdays, I feel sorry for them if I don’t reward them with some foods they like. Besides that, I don’t want them to experience peer pressure from their friends when comparing with each other.’ (SKB2-015); ‘It is hard for me to have control totally as the children will cook for themselves too. If they cook for themselves, I will reduce the times I cook that unhealthy food for them.’ (PPRK-005) ‘It is hard for me to have control totally as the children will cook for themselves too. If they cook for themselves, I will reduce the times I cook that unhealthy food for them.’ (PPRK-005)
3. Avoidance of unhealthy foods	Try to prolong the interval of giving to decrease the frequency of providing the foods Cook the foods craved by the child herself Decrease the amount of the foods bought and stored Educate the child about the side effects of taking those foods Cook/buy healthier choices of similar foods Distract the child’s attention while in conflict Check on pocket money and limit the pocket money of the children when not under the supervision Not giving those foods when under supervision Provide other foods as an alternative Ask the siblings to supervise each other while in school Punish the child when the child buys those foods themselves	‘I will still provide that food to my children but under control. Usually, I won’t buy too much of that food and not too frequent too.’ (SKB2-004) ‘Once in a month, when they request for that food, I will buy them similar food which is healthier. At times when they insist to buy the food, I will still control the amount they buy.’ (SKB2-007)

Key theme 2 (Reasons for providing unhealthy foods): Various reasons were reported by the participants when being asked why unhealthy foods were provided to their children. They were more likely to compensate or treat the child’s preferred food, i.e., ‘snack’ food as a reward. Besides that, some participants could not completely control their children’s diet. Their children would either buy or prepare unhealthy foods themselves when they were unsupervised by their parents. A few stated that other adults in the family could be barriers to controlling children’s food intake because they were the ones who provided unhealthy foods to the children. The participants mainly avoided confronting those adults.

Key theme 3 (Avoidance of unhealthy foods): Some participants reported that they would control the number of unhealthy foods purchased to reduce the frequency of providing such foods to the children. Some provided other foods as alternatives or cooked the foods in healthier methods. Other methods used by the participants such as educating the child about the side effects of excessive intake of those unhealthy foods, prolonging the interval of giving the foods to decrease the frequency of providing or distracting the child’s attention when they are conflicted about food choices. One participant reported that she had a family history of diabetes, so she avoided unhealthy foods exposure to her child completely to prevent them from getting diabetes. Through observation, her child was obese, which is a risk factor for diabetes. The awareness has created a restrictive feeding practice on the foods imposed on her children.

The FPQ used in Study 2 found across the 13 food groups, snacks were the most preferred food by the children [4.5 (0.0)] while legumes were the least preferred [2.0 (1.0)] as shown in Table 4. Within the snacks group, they preferred crispy snacks over the filled or flavoured biscuits with a slightly higher median preference score of 5.0 (1.0). Similarly, the parents scored the highest preference for the snacks group [4.5 (1.0)] and the lowest for legumes [2.0 (1.0)], hence no statistical difference was observed. Of the 36 individual food items listed, rice-based dishes, cultured and flavoured milk, and crispy snacks were the most preferred foods [5.0 (1.0)]. Table 5 outlines the factors affecting the food preferences as reported by the parent–child pairs. Based on the children’s responses, there were a total of nine key themes identified. Among all these factors, food sensory properties such as taste, texture, appearance and smell of the food appeared to be the most significant determinant of the children’s food preference. Moreover, some children also attributed their preference for certain foods to food association, food availability and accessibility, familial influence, internal cues, nutrition, health and wellbeing, media influence, as well as peer influence.

Table 4.

Number of children who tried each food and food preference score (study 2)

	Children			Parents
	Children who have tried the food		Food preference score	Children who have tried the food		Food preference score
Food item	N (%)	Median	IQR	N (%)	Median	IQR
Rice, noodle, bread, cereals, cereal products and tubers
Bread/pau/bun	14 (100)	4.0	1.0	14 (100)	4.0	1.0
Rice	14 (100)	4.0	1.0	14 (100)	4.0	2.0
Flour-based bread	14 (100)	4.0	2.0	14 (100)	4.0	1.0
Rice-based	14 (100)	5.0	1.0	14 (100)	5.0	1.0
Noodles-based	14 (100)	4.0	1.0	14 (100)	4.0	1.0
Fruits
Apple	14 (100)	4.0	1.0	14 (100)	4.0	2.0
Banana	14 (100)	4.0	1.0	14 (100)	4.0	1.0
Guava	13 (92.9)	4.0	3.0	14 (100)	3.0	2.0
Mango	14 (100)	5.0	1.0	14 (100)	4.0	2.0
Orange	14 (100)	4.0	1.0	14 (100)	4.0	0.0
Watermelon	14 (100)	4.0	3.0	14 (100)	4.0	2.0
Vegetables
Cabbage	13 (92.9)	3.0	1.0	14 (100)	2.0	1.0
Carrot	14 (100)	4.0	2.0	14 (100)	4.0	2.0
Cauliflower	13 (92.9)	2.0	3.0	14 (100)	3.0	2.0
Cucumber	13 (92.9)	4.0	2.0	14 (100)	3.0	3.0
Water spinach	12 (85.7)	3.0	2.0	14 (100)	3.0	2.0
Meat/poultry
Chicken	14 (100)	4.0	1.0	14 (100)	4.0	1.0
Hen egg	14 (100)	4.0	1.0	14 (100)	4.0	1.0
Processed meat/poultry	13 (92.9)	4.0	1.0	14 (100)	4.0	2.0
Fish and shellfish
Fish	14 (100)	4.0	2.0	14 (100)	4.0	2.0
Squid	13 (92.9)	4.0	2.0	14 (100)	4.0	3.0
Legumes
Chickpea/dhal/baked beans/green bean	12 (85.7)	2.0	2.0	14 (100)	2.0	1.0
Milk and milk products
Plain milk	14 (100)	5.0	1.0	14 (100)	4.0	2.0
Cultured milk	14 (100)	4.0	2.0	14 (100)	5.0	1.0
Flavoured milk	14 (100)	3.0	1.0	14 (100)	5.0	1.0
Sugar-sweetened beverages
Carbonated drinks	12 (85.7)	4.0	1.0	14 (100)	4.0	1.0
Sweetened beverages	13 (92.9)	5.0	1.0	14 (100)	4.0	2.0
Cocoa-based drink	14 (100)	4.0	1.0	14 (100)	4.0	2.0
Kuih, pastries, cakes
Steamed local kuih	12 (85.7)	4.0	2.0	14 (100)	3.0	2.0
Fried local kuih	14 (100)	4.0	2.0	14 (100)	3.0	1.0
Dessert	14 (100)	4.0	1.0	14 (100)	4.0	2.0
Snacks
Crispy snacks	14 (100)	5.0	1.0	14 (100)	5.0	1.0
Filled/flavoured biscuits	14 (100)	4.0	0.0	14 (100)	4.0	1.0
Sugars
Condensed milk	11 (78.6)	3.0	2.0	13 (92.9)	3.0	0.0
Oils
Cooking oil/mayonnaise/margarine/butter/salad dressing/gravy	13 (92.9)	3.0	2.0	14 (100)	3.0	1.0
Condiments
Sauces, vegemite	14 (100)	3.0	2.0	14 (100)	3.0	1.0

IQR, interquartile range.

Table 5.

Factors influencing children’s food preferences (self-reported and parental report)

Children self-report			Parental report
Key themes	Sub-themes	Responses	Key themes	Sub-themes	Responses
Food sensory properties	Taste (n = 14)	‘I dislike cabbage and cauliflower because they taste bitter.’ (C3-ADBAI) ‘I like steamed local kuih because it tastes sweet. And the shapes are beautiful.’ (C10-ABBMA); ‘I like water spinach because mummy always cooks it.’ (C10-ABBMA) ‘I like crispy snacks because they are easy to eat.’ (C3-ADBAI) ‘I like orange because it has vitamin C.’ (C9-TNE)	Food sensory properties	Taste (n = 13)	‘For my child, I think it is because of the taste of the food. The sweet taste will indeed cause her to like the food more.’ (P10-SBMS) ‘She likes the way I prepare it. If it is the Enoki mushroom, she will only take it in Tom Yam soup but no other cooking methods.’ (P07-NBAS) ‘To be honest, I think for the vegetables, it is because I introduced it very early when she was still young. You know, during the weaning stage?’ (P04-AAG) ‘I think one is because those foods are a part of our diet menu that we have provided to him since he was young.’ (P06-MNKEBH)
	Cooking method (n = 13)			Cooking method (n = 6)
	Texture (n = 13)			Texture (n = 5)
	Appearance (n = 3)			Appearance (n = 4)
	Smell (n = 2)
Food association	Food components/ingredients (n = 12)		Familial influence	Family exposure (n = 4)
	Food pairing (n = 9)			Family practice (n = 2)
	Food experiences (n = 2)			Family role modelling (n = 1)
Food availability and accessibility	Convenience (n = 6)		Food availability and accessibility	Convenience (n = 3)
	Store food availability (n = 6)			Food price (n = 2)
	School food availability (n = 5)			Cultural preference (n = 2)
Familial influence	Parental exposure (n = 7)		Internal cues	Craving (n = 6)
	Siblings’ exposure (n = 4)			Hunger (n = 3)
	Relatives’ exposure (n = 3)
Internal cues	Satiety (n = 8)		Media influence	Advertisements (n = 3)
	Appetite (n = 1)			Social media (n = 1)
Nutrition	Nutrition value (n = 5)		Meal pattern (n = 2)
	Energy supply (n = 4)
Health and Wellbeing (n = 7)			Food acceptance (n = 1)
Media influence	TV advertisements (n = 4)
	Social media (n = 1)
Peer influence (n = 3)

Likewise, the parents also perceived the main contributor to their children’s food preference to be the sensory properties. Furthermore, some parents attributed their children’s preference for certain foods to early introduction during infancy, family dietary practices, food culture and social media exposures.

DISCUSSION

To our best knowledge, these two studies were the first to explore parents’ perceptions of foods and practices affecting their feeding practices and children’s food preferences. Results from our studies extend previous literature examining parental feeding practices and dietary intake amongst Malaysian children [34–37]. In Study 1, the parent-reported child’s food preference was a major factor affecting their own feeding practices. Most parents did seek to find a balance between healthy food and beverages, and the child’s preference. Other parents masked nutritious foods like vegetables and fish using healthy cooking methods that were acceptable to the children such as soup or stir-fry. Such practices are usually adopted by parents of older children to ensure that their children have an adequate intake of healthy foods [47]. While this could be temporary, it provides an opportunity for parents to introduce healthy food to their children to learn to eat nutritiously by presenting children with one likable food and then gradually introducing similar foods to increase dietary variety (e.g. fried chicken to stir-fried chicken to stewed chicken to grilled chicken) [48]. Some strategies used by mothers such as prolonging the interval of providing unhealthy foods and educating about the side effects of excessive food intake could be counterproductive [49, 50]. Studies have shown that restrictive practices, a common parental feeding practice used in the traditional Asian culture [50], had significantly reduced children’s consumption of discretionary foods [49, 51]; however, its long-term effect should be examined closely to ascertain its effectiveness.

In Study 2, snacks were identified by the Malay parent–child pairs to be the most preferred food group by the children. From the parental perspective, food groups that were occasionally provided, such as snacks and sugar-sweetened beverages, were perceived as more preferred than core food groups. This is consistent with the evolutionary perspective, where discretionary foods/drinks could be over-consumed and driven by innate food preferences [19, 20]. One possible explanation for the least preferred food is the children’s unfamiliarity and lack of exposure to legumes in their food environment. This was consistent with findings from a local study reporting Malay primary school children exhibited a low preference for beans because legumes were less recognized by them [38]. Since repeated exposure is necessary to promote children’s familiarity and acceptability of food [52], leguminous vegetables would be less preferred as compared to the traditional vegetables such as ‘ulam’ in the Malay ethnicity, which the children had a higher preference for it [53].

The overall results in both studies showed that taste took precedence in children's food preferences similar to previous research on key determinants of children’s food preferences [21, 22, 54–57]. Of all the tastes, children generally showed a preference for sweet food, while expressing a dislike for bitter food, which is to the theory of biological influences on food preferences [22, 58]. For instance, sweetness may indicate a source of energy and calories that are essential for survival, while bitterness or sourness may signal the presence of toxins or bacteria that endanger health [58]. This explained why most of the children prefer fruits over vegetables and disliked beans the most. Other aspects such as food texture and appearance should be considered. While several studies found strong associations between preference for certain foods with their textures [38, 58], results were inconsistent for appearance which may be attributed to the use of plate colours, as opposed to food or packaging visuals [59].

The main limitation of both studies is the relatively small sample size, which cannot fully represent the feeding practices and food preferences among Malay children in Malaysia. Nevertheless, thematic data saturation was still achieved as further coding could not be conducted [45]. The homogenous sample of Malay participants residing in Klang Valley also restricts the generalizability of the research findings to other ethnic groups and geographical regions in Malaysia. Moreover, potential recall bias may arise throughout the data collection process as a participant-report method, rather than observational methods, was employed. Although breastfeeding plays a role in influencing children’s food preferences [60–63], it was not investigated in the present study due to its secondary effect compared to other environmental factors for children of this age group.

In conclusion, these two exploratory studies have provided key insights into children’s food intake and factors influencing these preferences. Overall, the knowledge of healthy and discretionary foods, and feeding practices are similar among the parents. Snacks were the most preferred food group while legumes were the least preferred food group reported by parent–child pairs. There were a variety of factors influencing these preferences, but food sensory properties and traditional Malay cuisine serving methods were the main contributors. The results provide a better understanding of the specific parental feeding practices, factors affecting the feeding practices, and barriers they faced when feeding their children. These factors may inform targeted strategies in food preparation and feeding practices to encourage healthy eating for families.

FUNDING

This research was funded by International Medical University [BDN I-2019 (20) and BDN I-2020 (12)] and supported by the University of Canberra Early Career Academic and Researcher Development Grant and University of Canberra Seed Grant.