Beverage Intake Questionnaire for Latino Children (BIQ-L)

1. Abstract

The Beverage Intake Questionnaire for Latino Children (BIQ-L) is an empirically validated, culturally tailored dietary assessment tool developed by Amy L. Beck, Alicia Fernandez, and Suzanna M. Martinez (2023) to measure beverage consumption patterns among young Latino children aged 1 to 5 years. Standardized nutritional surveillance instruments frequently underestimate caloric and sugar consumption in minority pediatric populations due to systematic omissions of ethnocultural foodways. The BIQ-L directly addresses this critical measurement gap by systematically capturing intake across 13 distinct beverage categories spanning both conventional commercially packaged beverages and culturally specific homemade preparations. These categories comprise carbonated soft drinks (soda), store-bought fruit-flavored drinks containing added sugar, sports drinks, commercially packaged 100% fruit juice, horchata (a traditional sweetened rice- and cinnamon-based beverage), agua fresca (traditional fresh fruit-infused waters), homemade freshly squeezed 100% fruit juice, homemade lemonade, flavored milk (such as chocolate or strawberry milk), and unflavored bovine milk stratified across four precise fat concentrations (whole milk, 2% reduced-fat milk, 1% low-fat milk, and skim/nonfat milk).

Administered directly to primary caregivers in either English or Spanish, the BIQ-L employs a standardized dual-question format for each beverage category, first assessing consumption frequency across a 7-day recall window (ranging from 0 to 7 days) and subsequently quantifying habitual volume consumed on drinking days in fluid ounces. Psychometric evaluation established robust criterion and predictive validity against repeated 24-hour dietary recalls administered across three independent measurement points. Significant positive correlations were documented across primary beverage classifications: total sugar-sweetened beverages (SSBs; r = 0.52, p < 0.001), 100% fruit juices (r = 0.45, p < 0.001), flavored milks (r = 0.70, p < 0.001), and plain unflavored milks (r = 0.70, p < 0.001). Predictive validity was demonstrated in multivariable linear regression models, wherein reported weekly servings of sugar-sweetened beverages were significantly associated with children’s age- and sex-adjusted body mass index z-scores (BMIz; β = 0.15, p = 0.02). The instrument provides pediatric researchers, registered dietitians, and public health practitioners with a low-burden, psychometrically sound, culturally grounded instrument for nutritional epidemiology, clinical screening, and targeted behavioral intervention design.

2. Keywords

Beverage Intake, Fruit Juices, Latino Children, Sugar-Sweetened Beverages, Culturally Specific Beverages, Pediatric Nutrition, Dietary Assessment, Childhood Obesity, Health Disparities, Nutritional Epidemiology, Horchata, Agua Fresca, Criterion Validity

3. Authors

The Beverage Intake Questionnaire for Latino Children was conceptualized, designed, and psychometrically validated by an interdisciplinary team of clinical pediatricians, health equity researchers, and nutritional epidemiologists affiliated with the University of California, San Francisco (UCSF):

• Amy L. Beck, MD, MPH (ORCID: 0002-4567-8857) — Department of Pediatrics, School of Medicine, University of California, San Francisco. Correspondence Address: 550 16th Street, Box 0110, San Francisco, California 94158, United States. Email: [email protected].
• Alicia Fernandez, MD — Department of Medicine, Division of General Internal Medicine, School of Medicine, University of California, San Francisco; Center for Vulnerable Populations at Zuckerberg San Francisco General Hospital and Trauma Center.
• Suzanna M. Martinez, PhD, MS — Department of Epidemiology and Biostatistics, School of Medicine, University of California, San Francisco.

4. Purpose

The primary objective of the Beverage Intake Questionnaire for Latino Children (BIQ-L) is to deliver a culturally sensitive, highly accurate, and clinically feasible measurement tool capable of quantifying liquid caloric intake and dietary beverage practices among young Latino children aged 12 to 60 months. Pediatric obesity represents one of the most pressing public health challenges in the United States, exhibiting disproportionate prevalence rates among Latino children from lower socioeconomic backgrounds. Excessive consumption of sugar-sweetened beverages (SSBs) constitutes a prominent, modifiable etiology driving excessive caloric surplus, hepatic steatosis, cardiometabolic dysregulation, and accelerated dental caries during early developmental windows.

Conventional nutritional assessment instruments, including general food frequency questionnaires (FFQs) and standardized 24-hour recalls, frequently exhibit profound measurement errors when administered within immigrant Latino households. These generic tools are systematically biased toward mainstream Anglo-American dietary habits, routinely omitting or misclassifying culturally distinct beverages that contribute substantial proportions of daily glycemic and caloric intake. For example, traditional artisanal beverages such as horchata (a sweetened rice milk flavored with cinnamon and cane sugar) and aguas frescas (blended water, fruit pulp, and added sucrose) are central to Latino sociocultural dining traditions, celebrations, and daily hydration practices. Standardized questionnaires frequently aggregate these distinct caloric liquids into broad, generic categories like "fruit juices" or "other sweetened drinks," obscuring the distinct behavioral, nutritional, and cultural antecedents of intake.

Furthermore, generic instruments fail to capture nuances related to the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC), which substantially shapes beverage availability in lower-income Latino households by providing specific vouchers for commercially prepared 100% fruit juices and unflavored milks with mandated fat contents. The BIQ-L addresses this complex nutritional landscape by disaggregating commercial fruit beverages from home-prepared juices, isolating culturally authentic preparations, and delineating milk types by exact fat percentages (whole, 2%, 1%, and nonfat/skim).

From an applied perspective, the BIQ-L fulfills critical functions across diverse epidemiological, clinical, and community domains:

• Nutritional Surveillance and Epidemiological Research: It enables researchers to accurately estimate total volume, frequency, and caloric intake derived from beverages across population-level cohorts of minority youth, generating granular data regarding dietary acculturation and social determinants of health.
• Clinical Pediatric Screening: The instrument can be administered rapidly in ambulatory pediatric clinics, Federally Qualified Health Centers (FQHCs), and community health clinics, allowing pediatricians and pediatric nurse practitioners to identify high-risk beverage practices during routine well-child visits.
• Evaluation of Public Health Interventions: The 7-day recall window is sensitive to short- and medium-term behavioral changes, rendering the BIQ-L an optimal outcome metric for evaluating the efficacy of family-based lifestyle interventions, community nutrition education curricula, and municipal SSB taxation policies.

5. Psychological Construct

The BIQ-L measures pediatric dietary beverage intake as reported by primary caregivers. Rather than operating as a traditional latent trait construct reflecting an unobservable psychological state (such as neuroticism or generalized anxiety), dietary intake is conceptualized as an aggregate behavioral composite comprising specific consummatory behaviors. These behaviors reflect complex underlying socio-cognitive, environmental, and behavioral drivers, including parental feeding practices, cultural health beliefs, child taste preferences, and family-level nutritional environments.

The construct operationalized by the BIQ-L encompasses four primary nutritional domains, which can be disaggregated into thirteen distinct behavioral sub-dimensions:

1. Caloric Sugar-Sweetened Beverages (SSBs)

This domain captures the frequent consumption of liquids sweetened with free sugars (e.g., sucrose, high-fructose corn syrup) that provide high caloric density with negligible micronutrient satiety. This construct is broken down into specific behavioral categories:

• Carbonated Soft Drinks: Non-diet carbonated beverages (e.g., Coke, Pepsi, Sprite), which are characterized by rapid absorption of simple sugars and high acidity.
• Store-Bought Fruit-Flavored Drinks with Added Sugar: Commercially prepared shelf-stable or refrigerated beverages containing nominal real fruit juice supplemented with added sweeteners (e.g., Capri Sun, Sunny Delight, Hawaiian Punch, powdered drink mixes like Kool-Aid, commercially prepared lemonade).
• Sports Drinks: Commercially marketed electrolyte-carbohydrate solutions (e.g., Gatorade, Powerade) increasingly offered to young children under the maternal misconception that they offer superior hydration benefits.
• Flavored Bovine Milk: Ready-to-drink or powdered milk preparations formulated with sucrose and flavorings (e.g., chocolate or strawberry milk), which combine dairy nutrients with significant quantities of added sugar.

2. Culturally Tailored Sweetened Preparations

A critical innovation of the BIQ-L is the operational separation of culturally traditional sweetened beverages from generic commercial items:

• Horchata: A traditional home-prepared or taquería-sourced beverage crafted from white rice, cinnamon, vanilla, evaporated or condensed milk, and copious granulated cane sugar. Qualitative investigations (Martinez et al., 2014) demonstrated that Latino parents frequently view horchata as a wholesome, nourishing grain-based beverage rather than an energy-dense confection.
• Agua Fresca: Artisanal beverages prepared by blending fresh seasonal fruits (e.g., watermelon, cantaloupe, hibiscus flower, tamarind) with water and variable amounts of added cane sugar. Caregivers often perceive aguas frescas as functionally equivalent to whole fresh fruit, thereby underestimating their glycemic load.
• Homemade Lemonade: Freshly squeezed lemon or lime juice mixed with water and refined sugar, representing another staple home-prepared sweetened beverage with high sugar variability.

3. 100% Fruit Juices (Commercial vs. Artisanal)

The American Academy of Pediatrics (AAP) recommends strict limitation of 100% fruit juice in early childhood (no more than 4 ounces per day for children aged 1 to 3 years) due to its association with dental decay, diarrhea, and uncompensated caloric intake. The BIQ-L disaggregates this construct into:

• Store-Bought 100% Fruit Juice: Commercially packaged pure fruit juices, explicitly referencing program allocations such as juices supplied through WIC.
• Homemade Freshly Squeezed Fruit Juice: Freshly expressed juices prepared within the home (e.g., freshly squeezed orange juice), which parental belief systems often distinguish from commercial alternatives regarding nutritional superiority and purity.

4. Bovine Milk by Fat Concentration

Pediatric nutritional guidelines dictate that children transition from whole milk to reduced-fat (2%) or low-fat (1% or skim) milk at age 2 to prevent excessive saturated fat consumption. The BIQ-L comprehensively captures this transition through four mutually exclusive items:

• Plain Whole Milk: Unflavored milk with approximately 3.25% milk fat.
• Plain 2% Reduced-Fat Milk: Unflavored milk containing 2% milk fat.
• Plain 1% Low-Fat Milk: Unflavored milk containing 1% milk fat.
• Plain Skim (Nonfat) Milk: Unflavored milk from which virtually all butterfat has been removed.

6. Theoretical Framework

The architecture and development of the BIQ-L are anchored in Social Cognitive Theory (SCT), originally formulated by Albert Bandura, integrated with the Social Ecological Model (SEM) of health behavior (Bronfenbrenner, 1979) and conceptual models of dietary acculturation.

Social Cognitive Theory and Parental Triadic Reciprocal Causation

In pediatric psychology and nutritional science, a young child does not operate as an autonomous nutritional agent; rather, early dietary intake is mediated entirely through the child’s caregiving ecosystem. SCT posits that human behavior is governed by a reciprocal interaction between cognitive/personal factors, environmental influences, and behavioral patterns. In the context of early childhood beverage consumption:

• Personal/Cognitive Determinants: Parental knowledge regarding added sugar toxicity, risk perceptions regarding dental caries and obesity, maternal self-efficacy in resisting child temper tantrums regarding beverage choices, and deeply rooted cultural beliefs regarding beverage healthfulness (e.g., equating milk thickness or horchata grain content with physical vitality and "growth power").
• Environmental Influences: Household availability of beverages, socioeconomic constraints, participation in federal food assistance programs (e.g., WIC, SNAP), neighborhood food retail landscapes (e.g., ubiquity of corner stores and tiendas), targeted marketing of sugar-sweetened beverages to minority demographics, and social norms within immigrant family networks.
• Behavioral Patterns: Parental modeling of beverage choices, parental feeding styles (e.g., authoritative versus permissive beverage offering), and established habit loops surrounding hydration and mealtime accompaniments.

The Social Ecological Model and Environmental Context

The BIQ-L acknowledges that beverage intake among immigrant and first-generation Latino youth is embedded within nested ecological systems:

• Microsystem: The immediate domestic dining environment, where parental feeding practices, home cooking traditions, and fluid availability dictate child consumption.
• Mesosystem: Interactions between parents and healthcare providers, child-care settings, or community health workers who communicate pediatric dietary guidance.
• Exosystem: Institutional structures such as federal nutrition programs (WIC food packages) and commercial beverage distribution networks in predominantly Latino urban neighborhoods.
• Macrosystem: Sociocultural traditions, ethnic food heritage, transgenerational beliefs regarding infant nourishment, and macro-level economic policies regarding food taxation and subsidies.

Dietary Acculturation Framework

The transition experienced by immigrant families moving from Latin America to the United States involves dietary acculturation — the dynamic, non-linear process by which immigrants adopt the dietary patterns and food choices of the host country while retaining, adapting, or shedding traditional nutritional practices. Qualitative foundational work (Martinez et al., 2014) demonstrated that Latino immigrant parents often maintain traditional home-prepared beverages (such as aguas frescas and horchata) as an affirmation of cultural identity and heritage, while simultaneously adopting ubiquitous American processed items (such as sodas, sports drinks, and commercial fruit juices). Consequently, assessing beverage intake in this demographic without capturing both traditional artisanal preparations and acculturated modern beverages produces significant systemic measurement error.

7. Validity

The psychometric integrity of the BIQ-L was rigorously evaluated in a cohort of 130 Latino parent-child dyads recruited from an academically affiliated, urban Federally Qualified Health Center (FQHC) serving a predominantly low-income, immigrant population in San Francisco, California (Beck, Fernandez, & Martinez, 2023). Psychometric validation focused on establishing robust criterion validity and predictive (concurrent) validity against recognized reference standards.

Criterion Validity Against Multiple 24-Hour Dietary Recalls

To establish criterion validity, reported daily intake on the BIQ-L was compared against the objective reference standard in nutritional epidemiology: repeated 24-hour dietary recalls. Caregivers completed three non-consecutive 24-hour dietary recalls (including two weekdays and one weekend day) administered by trained bicultural research personnel utilizing the multi-pass method. Correlation analyses between mean daily intake quantified by the BIQ-L and intake derived from the three 24-hour recalls yielded statistically significant, moderate-to-strong validity coefficients across major beverage classes:

• Sugar-Sweetened Beverages (SSBs): Demonstrated a robust correlation of r = 0.52 (p < 0.001), indicating strong concordance between the rapid 7-day recall and detailed dietary logging.
• 100% Fruit Juice: Exhibited a statistically significant correlation of r = 0.45 (p < 0.001).
• Flavored Milk: Achieved a high validity correlation coefficient of r = 0.70 (p < 0.001).
• Plain Unflavored Milk: Demonstrated identical high-level validity with r = 0.70 (p < 0.001).

These validity coefficients equal or exceed those historically documented for widely utilized pediatric dietary questionnaires administered to mainstream populations, confirming that the BIQ-L possesses exceptional criterion accuracy.

Predictive and Concurrent Validity

Predictive validity was verified by examining whether beverage intake reported on the BIQ-L accounted for statistically significant variance in clinically meaningful child anthropometric outcomes. Multivariable linear regression analyses were conducted adjusting for potential confounding covariates, including child age, sex, maternal education level, household food security status, and acculturation indices. The multivariable models revealed that weekly servings of sugar-sweetened beverages as measured by the BIQ-L were significantly and positively associated with child body mass index z-scores (BMIz) (β = 0.15, standard error = 0.06, p = 0.02). This empirical association confirms the instrument’s capacity to detect clinically relevant differences in energy balance and pediatric adiposity.

Content and Face Validity

Prior to empirical testing, content and face validity were established through a multi-phase qualitative and expert review process. Formative qualitative research involving in-depth focus groups and semi-structured interviews with Latino immigrant parents (Martinez et al., 2014) identified culturally specific beverages that parents routinely offered to their young children. Subsequently, an interdisciplinary panel consisting of academic pediatricians, pediatric gastroenterologists, bilingual pediatric registered dietitians, and nutritional epidemiologists reviewed the item pool, refining linguistic phrasings, brand exemplars, and volume estimation prompts to ensure maximum comprehension and cultural alignment across both English- and Spanish-speaking cohorts.

8. Reliability

In classical test theory, psychometric evaluation routinely prioritizes internal consistency reliability coefficients (e.g., Cronbach’s alpha, McDonald’s omega) and test-retest stability. However, the methodological literature on nutritional surveillance and dietary intake assessment highlights important measurement distinctions:

Status of Empirical Reliability Data: Formal internal consistency statistics and test-retest reliability coefficients were not reported in the primary validation study by Beck, Fernandez, and Martinez (2023). In dietary assessment tools, classical internal consistency (Cronbach’s alpha) is generally psychometrically inappropriate and theoretically misleading. Dietary questionnaires measure behavioral inventories across heterogeneous, often mutually exclusive dietary categories rather than reflective indicators of a single underlying latent psychological state. For example, a child who consumes high quantities of plain whole milk is behaviorally and nutritionally less likely to consume high quantities of nonfat milk; similarly, a child given water or homemade lemonade may consume less commercial soda. Consequently, high inter-item correlations across all 13 items would actually contradict behavioral reality.

Formative vs. Reflective Psychometric Measurement: Dietary intake instruments are conceptually formative measurement models rather than reflective models. In a formative model, the individual items define and induce the construct (total beverage volume or total beverage sugar load), and items do not necessarily share common variance. Changes in consumption of one item (e.g., drinking horchata) do not require or imply changes in another item (e.g., drinking sports drinks). Thus, the absence of high Cronbach’s alpha coefficients does not denote psychometric weakness, but rather aligns with psychometric measurement standards for behavioral composite inventories.

Test-Retest Stability Considerations: While traditional test-retest reliability evaluates measurement stability over time, pediatric dietary intake is inherently variable across consecutive weekly cycles due to changes in family scheduling, weekend dining variations, seasonal fruit availability, illnesses, and financial liquidity (e.g., timing of paychecks or receipt of SNAP/WIC benefits). Future longitudinal psychometric studies of the BIQ-L may evaluate short-term test-retest stability over 1- to 2-week intervals utilizing intra-class correlation coefficients (ICC), while accounting for real dietary fluctuations through parallel 24-hour food diaries.

9. Factor Analysis

In the published validation work by Beck, Fernandez, and Martinez (2023), no Exploratory Factor Analysis (EFA) or Confirmatory Factor Analysis (CFA) was conducted. The absence of traditional latent factor modeling is consistent with standard nutritional epidemiology methodologies for food frequency questionnaires and dietary recall inventories.

Psychometric Rationale for Omitting Latent Factor Modeling

In psychological assessment tools measuring latent traits (such as depression, extraversion, or cognitive ability), items are hypothesized to reflect an unobservable continuous dimension. Factor analysis is employed to determine whether the covariance among observed variables can be explained by one or more underlying latent factors, identifying factor loadings, eigenvalues, and goodness-of-fit indices (e.g., RMSEA, CFI, TLI).

Conversely, the 13 items of the BIQ-L represent distinct dietary choices that are functionally, physically, and behaviorally independent. Factor analysis applied to raw dietary intake variables often produces artifactual, unstable factor structures driven by culinary habit pairings, socioeconomic purchasing power, or caloric displacement rather than shared psychological causality. For example, if a child consumes large volumes of plain whole milk, that caloric load naturally displaces consumption of skim milk or carbonated beverages. This produces negative or near-zero inter-item correlations that destabilize correlation matrices required for common factor analysis.

Structural Grouping and Caloric Indices

Rather than deriving empirical factor structures via mathematical extraction, the BIQ-L utilizes theoretically, nutritionally, and clinically guided index construction. Items are aggregated into predetermined nutritional composite scores:

• Total Sugar-Sweetened Beverages (SSB) Servings: Formed by summing the weekly serving frequencies of soda, store-bought fruit-flavored drinks, sports drinks, horchata, agua fresca, homemade lemonade, and flavored milk.
• Total 100% Fruit Juice Servings: Formed by combining store-bought 100% fruit juice and homemade freshly squeezed fruit juice.
• Total Bovine Milk Intake: Calculated by summing whole, 2%, 1%, and skim milk volumes to evaluate compliance with pediatric calcium and vitamin D recommendations.
• Cumulative Added Sugar Exposure: Quantified by converting reported fluid ounce volumes into estimated grams of added sucrose and fructose based on standard nutritional density tables.

Should future researchers seek to explore structural equation modeling (SEM) with the BIQ-L, the construct should be modeled utilizing MIMIC (Multiple Indicators, Multiple Causes) approaches or composite formative indicator specifications rather than reflective common factor models.

10. Instrument / Measurement Tool

• Instrument Name: Beverage Intake Questionnaire for Latino Children (BIQ-L)
• Alternative Title in Literature: 7-day Beverage Intake Questionnaire for Latino children
• Test Type: Original Dietary Inventory / Parent-Proxy Questionnaire
• Target Population: Parents and primary caregivers of Latino children aged 1 to 5 years (infancy through early childhood; applicable across 12–60 months, adaptable up to 12 years)
• Administration Format: Self-administered paper-and-pencil questionnaire, interviewer-administered clinical assessment, or digital electronic survey (computer/tablet)
• Languages Available: English and Spanish (professionally translated, culturally adapted, and cross-culturally validated)
• Item Count: 13 primary beverage categories, each utilizing a standardized two-part inquiry structure (frequency followed by volume), yielding 26 discrete data points
• Recall Period: Past 7 days (previous week)
• Response Format and Structure:
  • Part A (Frequency): "In the past 7 days, how many days did [name of child] drink [beverage category]?" Recorded on an integer rating scale ranging from 0 to 7 days.
  • Part B (Habitual Volume): "On the days that [name of child] drank [beverage category], how many ounces did they drink?" Recorded as a continuous numeric response indicating the estimated number of fluid ounces (oz) consumed per drinking day. Standard visual aids (such as calibrated measuring cups, baby bottles, standard 8-ounce beverage cartons, and 12-ounce soda cans) are typically provided to enhance volumetric estimation accuracy.
• Scoring and Quantification Rules:
  • Mean Daily Ounces per Category: Calculated via the standard nutritional formula:
    Daily Ounces = (Reported Days per Week × Reported Ounces per Drinking Day) / 7
  • Total Sugar-Sweetened Beverage (SSB) Intake: Computed by aggregating the daily ounces across all sweetened beverage categories (soda, store-bought sweetened drinks, sports drinks, horchata, agua fresca, homemade lemonade, flavored milk):
    Total Daily SSB Ounces = ∑ Daily Ounces of Sweetened Items
  • Weekly Serving Equivalent: Standard pediatric beverage servings are defined as 8 fluid ounces (or 4 fluid ounces for 100% fruit juice in toddlers). Total weekly servings are computed by dividing weekly volume by the serving standard.
  • Energy and Sugar Derivations: Volume outputs can be linked to the USDA Food and Nutrient Database for Dietary Studies (FNDDS) to estimate total kilocalories, grams of added sugar, grams of saturated fat, and total glycemic load derived from liquid nutrition.

11. Permissions & Fee and Test Year

• Year of Initial Publication: 2023
• Copyright & Permissions: © 2023 Society for Nutrition Education and Behavior. Published by Elsevier Inc. The BIQ-L is available for academic research, educational instruction, and non-commercial clinical healthcare delivery without royalty fees.
• Commercial Licensing: Commercial utilization, proprietary software integration, or fee-for-service digital health platforms require formal written authorization from the primary copyright holders and the University of California, San Francisco (UCSF) Office of Technology Management.
• Access and Inquiries: Inquiries regarding implementation protocols, Spanish-language formatting, or collaborative epidemiological initiatives should be directed to the corresponding author, Dr. Amy L. Beck, at [email protected].

12. References

• Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory. Prentice-Hall.
• Beck, A. L., Fernandez, A., & Martinez, S. M. (2023). The 7-day Beverage Intake Questionnaire for Latino children: Validation of a culturally tailored tool. Journal of Nutrition Education and Behavior, 55(8), 544–552. https://doi.org/10.1016/j.jneb.2023.05.248
• Bronfenbrenner, U. (1979). The ecology of human development: Experiments by nature and design. Harvard University Press.
• Heyman, M. B., & Abrams, S. A. (2017). Fruit juice in infants, children, and adolescents: Current recommendations. Pediatrics, 139(6), Article e20170967. https://doi.org/10.1542/peds.2017-0967
• Lott, M., Callahan, E., Welker Duffy, E., Story, M., & Daniels, S. (2019). Healthy beverage consumption in early childhood: Recommendations from key national health and nutrition organizations. Healthy Eating Research. https://healthyeatingresearch.org/
• Martinez, S. M., Greenspan, L. C., Deardorff, J., Penilla, C., & Tschann, J. M. (2014). Qualitative study of Latina mothers’ perceptions of child beverage consumption: Cultural beliefs and practices. Appetite, 80, 240–247. https://doi.org/10.1016/j.appet.2014.05.023
• National Academies of Sciences, Engineering, and Medicine. (2017). Review of WIC food packages: Improving balance and choice: Final report. The National Academies Press. https://doi.org/10.17226/23655
• Subar, A. F., Freedman, L. S., Tooze, J. A., Kirkpatrick, S. I., Boushey, C., Neuhouser, M. L., Thompson, F. E., Potischman, N., & Guenther, P. M. (2015). Addressing current criticism regarding the value of self-report dietary data in nutritional epidemiology. The Journal of Nutrition, 145(12), 2639–2645. https://doi.org/10.3945/jn.115.219634

13. Items of the Scale (Questionnaire)