A food preference questionnaire (FPQ) is a structured survey that asks people to rate how much they like or dislike a list of specific foods, typically using a numbered scale. Researchers, clinicians, and food companies use these questionnaires to map patterns in what people enjoy eating, connect those patterns to health outcomes, and guide everything from dietary counseling to product development. The concept sounds simple, but the tools themselves have evolved considerably, with dozens of validated versions now tailored to different ages, cultures, and research goals.
What the Questionnaire Actually Looks Like
Most food preference questionnaires present a list of individual foods and ask respondents to rate each one. The most common format uses a 9-point hedonic scale, where 1 means “dislike extremely” and 9 means “like extremely,” with a neutral midpoint. This scale has been around since the 1950s and remains a workhorse in food science. When researchers compared the 9-point scale against two continuous alternatives (a Labeled Affective Magnitude scale and a Visual Analog Scale) using real berry varieties rated by the same group of people, the 9-point scale consistently performed as well or better at detecting differences in liking, and it reliably picked up differences at smaller sample sizes than the continuous scales.1Food Quality and Preference. A tale of 3 scales: How do the 9-pt, Labeled Affective Magnitude, and unstructured Visual Analog scales differentiate real product sets of fresh berries? Simpler 5-point versions also exist, and some researchers prefer ranking or best-worst scaling methods, especially when working with older adults who may find longer scales harder to use.2International Journal of Food Science & Technology. Comparison of discriminability of common food acceptance scales for the elderly
The food lists themselves vary enormously. Some questionnaires include 30 to 40 items, while others push past 70. The items are usually chosen to represent the dietary landscape of the population being studied, covering core food groups and common discretionary foods like sweets and snacks. When responses from large groups are analyzed statistically, clear clusters emerge. In a study of four-year-olds in the UK, preferences for 76 foods fell into four distinct groupings: vegetables, desserts, meat and fish, and fruit. Children who liked one vegetable tended to like others, independently of how they felt about desserts or meat.3PubMed. Factor-analytic structure of food preferences in four-year-old children in the UK A validation study in Iranian adolescents found a seven-factor structure, adding categories for dairy, snacks, starches, and miscellaneous foods to the mix.4PubMed Central. Validity and reliability of the Persian version of food preferences questionnaire (Persian-FPQ) in Iranian adolescents The number of factors depends partly on how many foods are tested and how diverse the diet is in a given culture, but the general principle holds: people’s preferences organize themselves into recognizable food-group clusters rather than random patterns.
Why Preferences Matter for Health Research
The practical value of knowing what someone likes becomes clear when you ask whether preferences predict what people actually eat. A study of young women found that food preferences alone explained about a quarter of the variation in fat consumption. Women who reported high preferences for meats and high-fat dairy products took in more of their calories from fat, while those who favored low-fat dairy ate less fat overall. Preferences for vegetables tracked with higher fiber intake, and preferences for fruit tracked with higher vitamin C intake.5The American Journal of Clinical Nutrition. Food preferences and reported frequencies of food consumption as predictors of dietary intakes in young women That same study found that for fiber and vitamin C, how often someone reported eating fruits and vegetables was a better predictor than preferences alone, but for fat consumption, knowing what someone liked was nearly as informative as knowing what they ate. This makes FPQs useful as a quick screening tool when a full dietary recall is impractical.
In obesity research, the connection between preferences and intake becomes especially relevant. A Japanese study using a newly developed nutrition-based FPQ found that people with abdominal obesity showed significantly greater preferences for foods high in carbohydrates, fat, and protein compared to non-obese participants. Within the obese group, there were also positive correlations between how much people liked high-fat and high-carbohydrate foods and how much of those foods they actually consumed.6PubMed Central. Food Preference Assessed by the Newly Developed Nutrition-Based Japan Food Preference Questionnaire and Its Association with Dietary Intake in Abdominal-Obese Subjects That kind of preference-intake link gives clinicians a way to identify eating patterns that may be driving weight gain, without requiring patients to keep detailed food diaries.
Measuring Children’s Preferences
Getting accurate data from children is one of the trickier challenges in this field. Young children cannot fill out paper surveys reliably, and their preferences shift as they encounter new foods. Parents are often used as proxies, but they do not always see eye to eye with their kids. A study of preschoolers using a computerized fruit and vegetable preference tool found that children’s own ratings were internally consistent and stable over time, with good test-retest reliability. But when researchers compared what children reported liking to what their parents said the children liked, the agreement was only moderate.7PubMed. Test-retest reliability and comparison of children’s reports with parents’ reports of young children’s fruit and vegetable preferences Parents tended to project their own sense of what their child enjoyed, which did not always match the child’s firsthand response.
For older children, self-reported tools work better. A pilot validation in Australian-Indian families developed two paired instruments: a Picky Eating Questionnaire completed by mothers, and a Child-reported Food Preference Questionnaire for children aged 7 to 12, covering 33 core food items and 11 discretionary foods.8Food Quality and Preference. The Picky Eating Questionnaire and Child-reported Food Preference Questionnaire: Pilot validation in Australian-Indian mothers and children 7-12 years old Using both questionnaires together lets researchers see where maternal perceptions of picky eating diverge from children’s actual preferences, which is valuable for understanding whether a child who seems picky at home genuinely dislikes certain foods or is responding to other social and environmental cues.
Genetics and the Biology of Taste
Food preferences are not purely learned. Your genes influence how intensely you perceive certain flavors, which in turn shapes what you gravitate toward or avoid. The most studied example involves the TAS2R38 gene, which codes for a bitter taste receptor. A systematic review of observational studies found a consistent correlation between variants of this gene and preferences for bitter and sweet foods.9PubMed Central. Genetic determinants of food preferences: a systematic review of observational studies In a cross-sectional study of Italian adults, people carrying a specific allele of TAS2R38 had higher thresholds for perceiving bitterness and showed distinctly different food preferences: they were roughly six times more likely to prefer beer, but less likely to favor butter or cured meat, compared to those with the other genotype.10PubMed. Association of the bitter taste receptor gene TAS2R38 (polymorphism RS713598) with sensory responsiveness, food preferences, biochemical parameters and body-composition markers
This matters for FPQ design because it means some preference patterns are partly hardwired. A questionnaire that captures someone’s general distaste for bitter vegetables is not just recording a quirk; it may be picking up on a biological trait with downstream health consequences. Researchers increasingly see FPQs as tools that can link genetic taste sensitivity to real-world dietary habits and, eventually, to metabolic outcomes. The early UK study of four-year-olds already noted that the factor structure of preferences could not be explained by simple sensory properties like sweetness or saltiness alone, suggesting more complex sensory and biological underpinnings.3PubMed. Factor-analytic structure of food preferences in four-year-old children in the UK
Clinical Uses Beyond General Nutrition
Food preference questionnaires also play a role in identifying and assessing eating disorders, particularly Avoidant/Restrictive Food Intake Disorder (ARFID). Unlike anorexia or bulimia, ARFID is not driven by body image concerns. People with ARFID may avoid food due to sensory sensitivity, lack of appetite, or fear of choking or vomiting. Screening tools built on food-preference principles help clinicians distinguish among these subtypes. A questionnaire version of the PARDI interview for ARFID identified three factors corresponding to the disorder’s known phenotypes and correctly flagged 90% of confirmed ARFID cases while correctly classifying 93% of healthy controls as unaffected.11PubMed Central. Preliminary validation of the pica, ARFID and rumination disorder interview ARFID questionnaire (PARDI-AR-Q) Population-level data from a large German sample estimated ARFID-like symptoms at just under 1% of adults, affecting men and women at similar rates.12PubMed. Psychometric evaluation of the Eating Disorders in Youth-Questionnaire when used in adults
Sensory-driven food avoidance is also common among children with neurodevelopmental conditions. Research comparing children with ADHD, Tourette Syndrome, autism spectrum disorder, and typical development has used parental reports of food fussiness, food preferences, and sensory sensitivity to tease apart the contributions of sensory processing to restrictive eating.13PubMed. The relationship between sensory sensitivity, food fussiness and food preferences in children with neurodevelopmental disorders For clinicians working with these populations, a food preference questionnaire is not just an academic exercise. It provides a structured way to figure out which foods a child will actually accept, which helps dietitians plan meals that meet nutritional needs without triggering avoidance behaviors.
How the Food Industry Uses Preference Data
Outside of health research, food companies rely heavily on preference measurement to guide product development. The plant-based food market is a clear example. A large European survey used a rapid sensory description method to map gaps between how consumers perceived existing plant-based alternatives to chicken, beef, cheese, yogurt, and milk versus how they imagined an ideal version of those products.14Food Quality and Preference. Consumers’ sensory-based cognitions of currently available and ideal plant-based food alternatives By identifying where current products fell short of consumer expectations on texture, flavor, and mouthfeel, the study gave manufacturers a concrete roadmap for reformulation. This kind of sensory preference mapping is essentially the same intellectual exercise as a clinical FPQ turned inside out: instead of asking “what does this person’s preference pattern mean for their health,” the question becomes “what does this population’s preference pattern mean for our product pipeline.”
Brain imaging research adds another layer. The Leeds Food Preference Questionnaire has been adapted for use alongside neuroimaging tools. In one study, Chinese adults completed the questionnaire while researchers tracked brain activity using functional near-infrared spectroscopy as participants viewed food images before and after eating a meal.15Food Quality and Preference. Neurobehavioral markers of food preference and reward in fasted and fed states and their association with eating behaviors in young Chinese adults Linking subjective preference ratings to measurable neural responses helps researchers understand how hunger, satiation, and reward processing influence what you say you want versus what your brain is actually responding to.
The Social Desirability Problem
One of the most persistent headaches with any self-reported food measure is that people bend the truth to look good, often without realizing it. This tendency, called social desirability bias, has been documented repeatedly in dietary research. In one landmark analysis, social desirability scores produced a downward bias in energy intake estimates of roughly 50 kilocalories per point on the desirability scale, adding up to about 450 kilocalories across the typical range of scores. The effect was about twice as large in women as in men, and people with the highest actual fat and energy intakes showed the biggest downward distortions.16PubMed. Social desirability bias in dietary self-report may compromise the validity of dietary intake measures
The bias does not just deflate totals. It selectively inflates “virtuous” foods and shrinks “guilty” ones. A study using food frequency questionnaires found that social desirability pushed vegetable consumption estimates upward for both men and women, while producing downward biases in reported intake of items like white bread, beer, and biscuits.17International Journal of Consumer Studies. Social desirability affects nutritional and food intake estimated from a food frequency questionnaire A randomized controlled trial confirmed that self-reports of fruit and vegetable intake from both food frequency questionnaires and 24-hour recalls were susceptible to substantial social approval bias, and the authors argued that valid assessments of dietary interventions may require objective measures of dietary change.18PubMed Central. Effects of social approval bias on self-reported fruit and vegetable consumption: a randomized controlled trial
For food preference questionnaires specifically, this means the data are cleaner when the question is genuinely about liking (“how much do you enjoy broccoli?”) rather than frequency of consumption (“how often do you eat broccoli?”). Liking a food carries less moral weight than claiming to eat it regularly, so the social desirability pressure is somewhat weaker. But it does not vanish entirely, and researchers designing FPQs increasingly build in checks or use indirect methods to account for it.
Do Preferences Stay Stable Over Time?
If your food preferences shifted dramatically from week to week, questionnaires built around them would be useless. The evidence suggests that at least some core preferences are quite stable. A study tracking women with overweight and obesity during a diet-induced weight-loss program found no significant changes in sweet food preferences over the course of the intervention. Participants maintained their relative position compared to others in the group throughout, with moderate to good stability for single measurements and good to excellent stability across multiple time points.19PubMed. Trait-like stability of sweet food preferences during diet-induced weight loss in women with overweight and obesity: Evidence from the Leeds food preference questionnaire That finding is reassuring for researchers: even during an active dietary intervention designed to change eating behavior, the underlying preference profile held steady, which supports treating sweet preference as a trait-like characteristic rather than a fleeting state.
This does not mean all preferences are fixed. Exposure to new foods, cultural shifts, aging, and medical treatments can all reshape what someone enjoys over months or years. But the short-term reliability is strong enough that a single questionnaire administration provides a meaningful snapshot of where someone stands.
Photo-Based and Digital Approaches
Technology has opened up new ways to collect preference data. Photo-based online questionnaires let participants rate images of foods rather than reading text descriptions, which can be helpful for children or people with low literacy. But images introduce their own problems. A study that compared online photo-based ratings to in-lab taste tests found that predicted liking from photos explained only about 19% of the variation in actual tasted liking for familiar snacks and about 35% for unfamiliar ones. Prospective consumption ratings (how much someone said they would eat) did even worse, explaining only about 5 to 11% of what people actually consumed in the lab.20Journal of Sensory Studies. Using an online photo based questionnaire to predict tasted liking and amount sampled of familiar and unfamiliar foods by female nutrition students The researchers described the correlation as poor, which is a useful reality check: what you think you will enjoy when looking at a picture and what you actually enjoy when tasting the food are only loosely related.
Digital photography has had more success when used to measure what people are actually eating rather than predicting what they would like. Comparing photographs of served and leftover food against weighed portion sizes showed that the photo method yielded estimates with a bias of less than 1.5 grams, comparable to visual estimation by trained observers.21PubMed. Comparison of digital photography to weighed and visual estimation of portion sizes So digital tools work well for tracking intake but are limited when it comes to predicting preferences from images alone.
Adapting Questionnaires Across Cultures
A food preference questionnaire built for one country’s diet cannot simply be translated word for word and used elsewhere. Foods that are staples in one culture may not exist in another, and even shared foods carry different flavor expectations. Cross-cultural adaptation typically follows a structured process that includes translation, back-translation, and pilot comprehension testing to ensure the adapted version measures the same constructs.22PubMed Central. Cross-Cultural Adaptation, Validity, and Reproducibility of the Mediterranean Islands Study Food Frequency Questionnaire in the Elderly Population Living in the Spanish Mediterranean The Persian version of the FPQ validated in Iranian adolescents, for example, arrived at seven factors rather than the four found in British preschoolers, partly because the adolescent Iranian diet includes a wider variety of starches, dairy products, and snack foods that warranted their own categories.4PubMed Central. Validity and reliability of the Persian version of food preferences questionnaire (Persian-FPQ) in Iranian adolescents
This means that comparing preference data across populations requires caution. A “high vegetable preference” score in one culture might represent enthusiasm for leafy greens and root vegetables, while in another it reflects liking for salads and stir-fried dishes. The underlying constructs are broadly similar, but the specific foods loading onto each factor shift with the local diet. Researchers who ignore this end up comparing apples to, well, something that is not an apple in the other country’s food system. For multinational studies, this adaptation step is not optional; it is what makes the data interpretable at all.