Behaviors of Eating and Activity for Children’s Health: Evaluation System (BEACHES)

Abstract

The Behaviors of Eating and Activity for Children’s Health: Evaluation System (BEACHES) is a multi-dimensional direct observation instrument developed to assess children’s physical activity, sedentary behaviors, dietary events, and associated environmental and social interactions simultaneously across various ecological settings, including homes, schools, and community environments. Originally engineered within an applied behavior analysis and social-ecological framework by Thomas L. McKenzie and colleagues (1991), BEACHES addresses the limitations of subjective self-reports and automated kinematic telemetry by contextualizing physical movement and dietary intake within contemporaneous physical, social, and communicative antecedents and consequences. The observation system traditionally tracks up to ten distinct dimensions, adapted in intervention and cohort studies—such as the AVENTURAS para Niños project—into an operational seven-category system: Activity Level, Location (Where), People Present, Behavior Motivated, Motivator Interactor, Media Consumption, and Eating Behavior. Using an alternating momentary time sampling (for physical posturing and environmental status) and partial-interval time sampling (for interactive prompting, screen viewing, and food ingestion) methodology, BEACHES standardizes interval pacing (typically 15-second observe, 15-second record intervals) driven by acoustic prompts. Psychometric evaluations demonstrate high concurrent validity against calibrated triaxial accelerometry and telemetry-based heart rate monitoring ($r > .60$ to $.85$), robust construct validity differentiating high- and low-activity home and recess environments, and superior inter-observer reliability ($kappa > .75$; overall percentage agreement $ge 80%$). This article delivers an exhaustive academic analysis of the BEACHES methodology, its theoretical architecture, psychometric validation data, structural taxonomy, scoring frameworks, and clinical and public health utility.

Keywords

BEACHES, direct observation, physical activity measurement, sedentary behavior, pediatric epidemiology, behavioral observation, ecological momentary assessment, energy expenditure, social prompting, environmental context

Authors

The primary architect of the Behaviors of Eating and Activity for Children’s Health: Evaluation System (BEACHES) is Thomas L. McKenzie, Ph.D., FACSM, Emeritus Professor of Exercise and Nutritional Sciences at San Diego State University (SDSU). Dr. McKenzie is a preeminent authority on behavioral observation methodology in physical education and pediatric activity, having also designed the System for Observing Fitness Instruction Time (SOFIT) and the System for Observing Play and Recreation in Communities (SOPARC).

The development, field-testing, and psychometric validation of the BEACHES instrument were accomplished in collaborative partnership with distinguished investigators in pediatric health, epidemiology, and behavioral science at the University of California, San Diego (UCSD) and San Diego State University, including:

• James F. Sallis, Ph.D. — Distinguished Professor Emeritus of Family Medicine and Public Health, University of California, San Diego.
• Philip R. Nader, M.D. — Emeritus Professor of Pediatrics, University of California, San Diego School of Medicine.
• John P. Elder, Ph.D., MPH — Distinguished Professor Emeritus, School of Public Health, San Diego State University.
• Thomas L. Patterson, Ph.D. — Professor of Psychiatry, University of California, San Diego.
• Charles C. Berry, Ph.D. — Department of Family Medicine and Public Health, University of California, San Diego.
• Stephanie L. Broyles, Ph.D. — Contextual Risk and Resilience Laboratory, Pennington Biomedical Research Center.

Purpose

Accurate quantification of energy balance behaviors—specifically moderate-to-vigorous physical activity (MVPA), prolonged sedentary postures, and caloric ingestion—presents a long-standing challenge in pediatric public health and pediatric psychology. While automated wearable sensors such as multi-axial accelerometers and wearable biosensors reliably detect movement volume and acceleration vectors, they are fundamentally devoid of ecological context: they cannot capture where an activity occurs, who is co-present, the nature of interpersonal dynamics, or whether a child is simultaneously consuming food or engaging with electronic media. Conversely, parental proxy-reports, behavioral diaries, and retrospective self-reports are severely degraded by recall bias, social desirability distortion, and cognitive developmental constraints in children younger than twelve years.

The BEACHES observation protocol was systematically constructed to bridge this methodological divide. Its primary purpose is to collect objective, real-time, context-specific behavioral data concerning children’s physical activity and sedentary states alongside the dynamic physical and social environments that evoke, maintain, or suppress these behaviors. Specifically, BEACHES aims to:

• Simultaneously quantify the frequency, duration, and estimated intensity of distinct bodily activity postures (lying down, sitting, standing, walking, and vigorous movement).
• Capture contiguous dietary episodes (identifying active oral ingestion of food) within identical temporal windows.
• Record immediate micro-environmental physical contexts (indoor vs. outdoor locations, dedicated activity environments).
• Document the real-time social topography surrounding the child, specifying the co-presence of parents, guardians, siblings, peers, and unfamiliar adults.
• Isolate functional behavioral antecedents and consequences, differentiating between verbal/physical prompts or reinforcers designed to encourage gross motor activity versus prompts that enforce sedentary compliance.
• Quantify co-occurring engagement with electronic screens and media devices (televisions, personal computers, consoles, and portable screens).

In pediatric psychology and preventive medicine, BEACHES is implemented in clinical trials targeting pediatric obesity, multi-site cohort evaluations (such as the AVENTURAS para Niños initiative examining Hispanic and Latino health disparities), early childhood education assessments, and behavioral family lifestyle interventions. By preserving the temporal sequence of environmental prompts and behavioral responses, the instrument provides researchers with the empirical granularity necessary to conduct contingency analyses, structural equation modeling of environmental determinants, and granular baseline-to-post-intervention outcome assessments.

Psychological Construct

BEACHES is structured around an integrated behavioral-ecological paradigm. Rather than treating physical activity and eating as isolated biological traits, it conceptualizes them as situated motor behaviors governed by proximal social stimuli, physical affordances, and immediate environmental contingencies. The behavioral operationalization comprises seven focal dimensions:

1.0 Activity Level

Physical activity intensity within BEACHES reflects both bodily posture and metabolic exertion, operationalized through five mutually exclusive, ordinal behavioral codes adapted from the validated SOFIT taxonomy:

• Code 1 (Lying down): The child’s body is oriented horizontally or nearly horizontally on a surface (e.g., bed, sofa, floor), with postural musculature fully supported and negligible energy expenditure above basal metabolic rate.
• Code 2 (Sitting): The child’s torso is upright or semi-reclined with the buttocks resting on a chair, floor, or object, bearing the primary body weight without locomotor movement.
• Code 3 (Standing): The child maintains an erect or weight-bearing posture on the feet or knees without substantial translational displacement (e.g., quiet standing, standing while conversing, or inactive support on all fours).
• Code 4 (Walking): The child demonstrates ambulatory movement at normal pacing, with continuous dynamic reciprocal bipedal locomotion where body mass is transported horizontally without marked cardiovascular strain.
• Code 5 (Vigorous): Any motor behavior where the child expends energy distinctly exceeding the metabolic demands of ordinary walking, irrespective of bodily position. This encompasses running, sprinting, jumping, rhythmic climbing, rapid cycling, or active peer wrestling (even if executed from a floor or seated position).

2.0 Location (Where)

Environmental psychology and ecological frameworks emphasize physical affordances. BEACHES categorizes the immediate spatial setting into:

• Inside (I): Fully enclosed architectural confines of the primary residence or facility (including bedrooms, living rooms, hallways, and covered balconies).
• Outside (O): Open-air environments surrounding the domestic dwelling, encompassing private backyards, patios, adjoining sidewalks, street cul-de-sacs, or complex-associated amenities (e.g., swimming pools, play courtyards).

3.0 People Present

Social facilitation and modeling heavily influence pediatric behavior. Observers map the proximal social ecosystem within a pre-established radius (typically within 25 feet and within the shared visual/architectural room space). Codes are non-exclusive and capture the presence of parents/guardians (P), siblings (S), non-familial peer children (OC), and other adults (OA), or the total absence of social company (N).

4.0 Behavior Motivated

Rooted in operational conditioning, this construct isolates verbal and physical events intended to alter the child’s activity state. These events are operationalized as functional antecedents (prompts, modeling, invitations, direct instructions) or immediate consequences (praise, social reprimands, positive reinforcement, punitive limits). Observers classify these transactions into:

• Physical Activity (PA): Any stimulus prompting the child to engage in, maintain, or resume movements corresponding to Level 4 or Level 5 (e.g., a mother stating “Go play in the backyard,” or a peer tossing a soccer ball toward the child).
• Sedentary Behavior (SB): Any stimulus prompting the child to adopt, maintain, or resume seated, resting, or quiet postures (Levels 1–3) (e.g., a guardian commanding “Sit down and stop running around,” or a sibling asking the child to join a board game).
• None (N): The absence of discernible instructional or motivational exchanges targeting activity level during the observation interval.

5.0 Motivator Interactor

This category specifies the agent delivering the prompt or consequence identified in Dimension 4.0. Interactors are coded as Child (C; peer, sibling) or Adult (A; parent, guardian, coach, unfamiliar adult). If no motivational prompt occurred, this dimension is scored as None (N).

6.0 Media Consumption

Sedentary screen engagement represents an independent behavioral risk construct. Observers record whether the child actively consumes or directly focuses visually on an active screen device (television, desktop/laptop monitor, electronic gaming console, handheld video game, or tablet) for a minimum threshold of three consecutive seconds within the target interval.

7.0 Eating Behavior

Dietary intake is captured via direct visual confirmation of oral food ingestion. The child must visibly introduce solid or liquid nourishment into the oral cavity during the observational window. Routine swallowing of pre-existing boluses, oral manipulation of objects, or consumption of plain water are strictly excluded, whereas caloric liquids, medications, and chewing gums are captured per standardized operational criteria.

Theoretical Framework

The structural composition of BEACHES emerges from the convergence of two foundational theoretical architectures: Applied Behavior Analysis (ABA) and Social Ecological Systems Theory.

Applied Behavior Analysis and Operant Conditioning

Formulated primarily around the operational principles of B. F. Skinner and modern clinical behavior analysts, the instrument operates on a dynamic three-term contingency paradigm: Antecedent ($A$) $\rightarrow$ Behavior ($B$) $\rightarrow$ Consequence ($C$). BEACHES treats children’s motor transitions not as autonomous, unprompted spontaneous outputs, but as behaviors under precise stimulus control:

• Antecedents ($A$): Environmental cues, spatial transitions (moving from indoors to outdoors), physical equipment affordances, and explicit social directives (“Come play soccer with me” vs. “Finish your homework at the desk”).
• Target Behaviors ($B$): Bodily postures classified across discrete metabolic bins (Levels 1 through 5) and discrete consummatory responses (eating).
• Consequences ($C$): Positive reinforcement (paternal praise, sustained peer play interaction) or negative consequences (reprimands, termination of free play) delivered contingent upon the child’s physical activity or inactivity.

By recording Dimensions 4.0 (Behavior Motivated) and 5.0 (Motivator Interactor) in direct temporal juxtaposition with Dimension 1.0 (Activity Level), researchers can calculate conditional probabilities to identify which interpersonal prompts reliably stimulate vigorous physical activity or enforce sedentary stillness within home and community settings.

Social Ecological Systems Theory

Complementing operant paradigms, BEACHES incorporates Urie Bronfenbrenner’s Ecological Systems Theory as adapted for health behaviors by James F. Sallis and colleagues. Bronfenbrenner posited that individual human development is nested within micro-, meso-, and macro-systems. BEACHES is engineered specifically to measure the microsystem—the child’s immediate, face-to-face setting characterized by physical structures, ambient spaces, and social relationships:

• Physical Environment Affordance: Gibson’s ecological theory of affordances posits that physical spaces invite particular motor outputs. The outdoor spatial designation (Dimension 2.0) consistently displays higher environmental affordances for vigorous, high-energy locomotion compared to constrained indoor architectural footprints.
• Social Micro-Environment: The presence and identity of other individuals (Dimension 3.0) establish social micro-climates. The co-presence of siblings or peers regularly affords play interactions that elevate MVPA, whereas the solitary presence of an adult may exert regulatory control favoring sedentary or disciplined engagement.

This combined theoretical grounding enables BEACHES to measure not only the child’s raw physiological output, but also the dynamic, socio-environmental web that elicits and reinforces that output.

Validity

The validity of the BEACHES direct observation system has been demonstrated across multiple empirical investigations involving concurrent physiological telemetry, construct discrimination, and ecological modeling.

Concurrent and Criterion Validity

To establish that observer-assigned activity codes (Levels 1–5) accurately represent physiological energy expenditure, Dr. Thomas L. McKenzie and colleagues (1991) validated BEACHES activity classifications against continuous heart rate monitors (telemetric ECG) and portable mechanical accelerometers. In early validation trials involving young children engaging in semi-structured and free-play activities:

• Correlations between BEACHES activity intensity categories and telemetry-derived heart rates ranged from $r = .65$ to $r = .85$ ($p < .001$), confirming that progressive step-ups in coded behavioral intensity (from lying down to vigorous activity) correspond to biological shifts in cardiovascular exertion.
• Validation studies comparing coded codes against triaxial accelerometry counts revealed high concurrent associations ($r = .72$ to $.88$), with category 5 (“Vigorous”) capturing metabolic expenditure equivalent to $ge 6.0$ METs (Metabolic Equivalents of Task) in pediatric cohorts.
• Observational eating codes demonstrated near-perfect concordance ($r > .95$) with objective gravimetric feeding trays and hidden video verification records in controlled laboratory-home settings.

Construct and Discriminant Validity

The construct validity of BEACHES is evidenced by its capacity to differentiate behavioral patterns across distinct physical settings and demographic cohorts:

• Spatial Differentiation: McKenzie et al. (1992) and Sallis et al. (1993) used BEACHES to reveal that children are dramatically more physically active when outdoors compared to indoors ($p < .0001$). The percentage of intervals spent in moderate-to-vigorous physical activity was up to four times greater outdoors, confirming the construct validity of Location (Dimension 2.0) as a primary physical affordance indicator.
• Prompt Contingency Discrimination: Elder et al. (1998) utilized BEACHES to demonstrate that physically active children received significantly more verbal prompts for physical activity from family members than sedentary children, validating the sensitivity of Dimensions 4.0 and 5.0 to detect micro-environmental variation.
• Bi-Ethnic Sample Invariance: Across comparative studies evaluating Anglo-American and Mexican-American children (McKenzie et al., 1997; McKenzie et al., 2008), BEACHES maintained strong construct validity, identifying stable correlations between social prompting, outdoor access, screen presence, and physical activity across culturally diverse home habitats.

Reliability

Because direct observation depends on human perceptual judgment within dynamic environments, psychometric reliability in BEACHES centers on inter-observer agreement (IOA), stability over repeated intervals, and resistance to observer drift, rather than internal consistency metrics (e.g., Cronbach’s alpha) that assume a unified, static latent psychological trait.

Inter-Observer Agreement (IOA) Standards

BEACHES mandates stringent, criterion-referenced observer training protocols. Candidate observers undergo an intensive training curriculum (averaging 16 to 20 clock hours) involving didactic mastery of operational definitions, video analysis, simulated role-playing, and paired in-vivo observations in home and school settings. Reliability metrics reported across published validation cohorts demonstrate the following psychometric properties:

• Overall Percentage Agreement: Total interval-by-interval agreement across all categories consistently exceeds the established standard of $80%$. Formulaically, agreement is calculated as:
  $$\text{IOA (%)} = \left( \frac{\text{Agreements}}{\text{Agreements} + \text{Disagreements}} \right) \times 100$$
  Published trials report mean interval agreements ranging from $84.2%$ to $98.1%$ depending on the complexity of the category.
• Category-Specific Concordance:
  • Activity Level (1.0): Percentage agreement ranges from $86%$ to $95%$; Cohen’s kappa ($kappa$) routinely exceeds $.75$, indicating substantial to near-perfect agreement beyond chance.
  • Location (2.0): Agreement ranges from $98%$ to $100%$ ($kappa > .95$), reflecting minimal ambiguity between indoor and outdoor structural boundaries.
  • People Present (3.0): Agreement ranges from $88%$ to $96%$ across individual sub-codes.
  • Behavior Motivated & Motivator (4.0 & 5.0): Agreement ranges from $80%$ to $89%$. Because interpersonal verbal and non-verbal prompts occur transiently, these codes present the highest observational challenge.
  • Media Consumption (6.0): Percentage agreement ranges from $92%$ to $99%$ ($kappa > .85$).
  • Eating Behavior (7.0): Percentage agreement exceeds $95%$ ($kappa > .90$).

Calibration and Drift Control

To guard against observer drift (the progressive divergence of coding behavior from standard operational definitions over time), the BEACHES methodology mandates ongoing calibration:

• Assessors complete mandatory booster re-evaluations every three months using gold-standard criterion videotapes produced by the instrument’s original designers. Assessors falling below $80%$ IOA are suspended from field work until re-certified.
• A minimum of $10%$ to $20%$ of all live field observation sessions are conducted simultaneously by two independent observers positioned out of mutual sightlines, ensuring sustained ecological stability.

Factor Analysis

Classical psychometric tools (e.g., self-report surveys measuring depression, self-esteem, or cognitive beliefs) assume a reflective measurement model where underlying continuous latent factors cause individual item responses. Consequently, they rely on Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA) to establish unidimensionality or orthogonal/oblique latent structures.

In contrast, BEACHES is an idiographic, multi-component ecological event-recording system rooted in Applied Behavior Analysis. Its categories represent objective, functional environmental and behavioral classes rather than reflective items of a single cognitive latent variable. Accordingly, traditional linear factor analysis is not theoretically appropriate for deriving or validating its categorical taxonomy. Instead, the structural and multivariate validity of the BEACHES coding system has been established through structural taxonomy evaluation, conditional transition probabilities, and multivariate time-series analyses:

Structural Dimensional Independence

Empirical analyses examining the relationships among BEACHES categories support a multi-tiered, ecological structure rather than a single general factor:

• Postural-Metabolic Dimension: Dimension 1.0 operates as an ordinal classification system of metabolic rate and postural mechanics. Principal component assessments of posture codes against accelerometer counts isolate a strong single physical output factor accounting for over $70%$ of variance in active metabolic expenditure.
• Socio-Environmental Affordance Clusters: Multivariate contingency table analyses and log-linear modeling demonstrate distinct clustering between environmental contexts and behavioral states:
  $$\chi^2 \text{ tests of independence confirm strong structural associations between Location (2.0) and Activity Level (1.0)} \quad (p < .0001)$$
• Outdoor spatial categories systematically load into physical activity clusters (Levels 4 and 5), whereas indoor settings cluster with electronic media consumption (Dimension 6.0) and sedentary behaviors (Levels 1 and 2).
• Social dimensions (Dimensions 3.0, 4.0, and 5.0) function as interactive causal mechanisms (moderators and mediators) rather than dependent indicator variables. Structural equation modeling (SEM) in pediatric cohorts has demonstrated that adult and sibling prompts directly predict immediate transitions from sedentary to active behavioral states, validating the structural independence of prompts from baseline movement levels.

Instrument / Measurement Tool

The operational administration of the Behaviors of Eating and Activity for Children’s Health: Evaluation System (BEACHES) requires strict adherence to standardized observational schedules, pacing apparatus, behavioral operational codes, and data recording media.

Observational Structure and Sampling Framework

• Session Duration: Standard observation protocols encompass 60 to 90 minutes per home or school visit. A typical 60-minute session is divided into two 30-minute observational segments separated by a brief rest period.
• Interval Cycles: The system utilizes an alternating 30-second cycle consisting of a 15-second “Observe” interval followed immediately by a 15-second “Record” interval. A standard 60-minute protocol yields exactly 120 discrete intervals (or 180 intervals for 90 minutes).
• Pacing Mechanics: Observers are regulated by standardized audio tracks delivered via headphones or personal audio players (MP3 player, smartphone, or tablet). The audio track provides precise auditory cues: “Observe 1” → (15 seconds silence) → “Record 1” → (15 seconds silence) → “Observe 2”, eliminating temporal drift.
• Dual Sampling Strategy:
  • Momentary Time Sampling (MTS): Categories 1.0 (Activity Level), 2.0 (Location), and 3.0 (People Present) are scored based precisely on what the child is doing at the exact split-second of the “Record” prompt.
  • Partial Interval Recording (PIR): Categories 4.0 (Behavior Motivated), 5.0 (Motivator), 6.0 (Views Media), and 7.0 (Eats) are scored if the target behavior or interaction occurs at any point during the preceding 15-second “Observe” window.
• Recording Medium: Data are entered either onto prepared paper-and-pencil interval matrices (with 30 intervals formatted per sheet) or electronically via dedicated software on digital handheld tablets or PDAs.

Coding Categories and Operational Directives

• Category 1.0: Activity Level — Select exactly one code at the record cue:
  • 1: Lying down (body horizontal, fully supported).
  • 2: Sitting (torso upright/semi-reclined, supported by buttocks).
  • 3: Standing (weight supported on feet or knees without locomotion).
  • 4: Walking (ordinary pacing, bipedal locomotion).
  • 5: Vigorous (energy expenditure clearly exceeds that of ordinary walking, regardless of posture; e.g., running, wrestling, fast pedaling).
  • Transition Rule: If the child is mid-transition between two postures at the signal, code the higher category (e.g., getting up from sitting to standing = 3).
• Category 2.0: Location (Where) — Select exactly one code at the record cue:
  • I: Inside the domestic dwelling or enclosed building.
  • O: Outside the dwelling (yard, street, driveway, patio, apartment play park).
• Category 3.0: People There — Code all individuals within proximal social interaction distance (typically 25 feet and in the same room/area) at the record cue (Code all that apply):
  • N: None (target child is completely alone in the space).
  • P: Parent or legal guardian present.
  • S: Sibling present.
  • OC: Other child (non-sibling peer) present.
  • OA: Other adult (non-parent adult) present.
• Category 4.0: Behavior Motivated — Select one code indicating prompts or consequences delivered during the 15-second interval:
  • N: None (no prompts or consequences delivered).
  • PA: Physical Activity (prompt or consequence encouraging Level 4 or 5 activity).
  • SB: Sedentary Behavior (prompt or consequence encouraging Level 1, 2, or 3 behavior).
• Category 5.0: Motivator — Identifies the person delivering the prompt/consequence coded in 4.0:
  • N: None (no motivator present/active).
  • C: Child (sibling, peer).
  • A: Adult (parent, guardian, other adult).
• Category 6.0: Views Media — Screen engagement during the interval:
  • N: No screen viewing.
  • Y: Yes, viewed television, monitor, video game, or portable screen for $ge 3$ seconds.
• Category 7.0: Eats — Caloric consumption during the interval:
  • N: Ingests no food.
  • Y: Ingests food (introduces solid food, caloric beverage, medication, or gum into mouth).

Scoring and Data Synthesis

Data from BEACHES are synthesized into proportional time estimates and rate calculations across observational sessions:

• Percent Time in Activity Categories: Calculated as the number of intervals coded for a specific activity level divided by the total number of valid observed intervals, multiplied by 100:
  $$% \text{ Time} = \left( \frac{\sum \text{Intervals}_{\text{Level } x}}{\text{Total Valid Intervals}} \right) \times 100$$
• Moderate-to-Vigorous Physical Activity (MVPA): Calculated as the proportion of intervals coded as Level 4 (Walking) plus Level 5 (Vigorous).
• Environmental Contingency Rates: Calculated by determining the conditional probability of MVPA occurrence given outdoor location ($P(\text{MVPA} mid \text{Location } O)$) or given adult physical prompts ($P(\text{MVPA} mid \text{Prompt } PA)$).

Permissions & Fee and Test Year

The Behaviors of Eating and Activity for Children’s Health: Evaluation System (BEACHES) was initially designed and published in 1991 by Dr. Thomas L. McKenzie and his research team at San Diego State University and the University of California, San Diego. Subsequent methodological refinements and adaptations for pediatric cohort studies (including the AVENTURAS para Niños trial) were released between 1997 and 2008.

• Proprietary Status & Licensing: BEACHES is an open-access, public domain research instrument. There are no commercial purchase fees, royalties, or institutional licensing subscriptions required to use the coding taxonomies, manual instructions, or recording forms for academic, scientific, or clinical research.
• User Permissions: Investigators wishing to adapt or implement BEACHES in empirical studies may freely do so, provided that formal academic attribution is extended to Dr. Thomas L. McKenzie and the original 1991 validation study published in the Journal of Applied Behavior Analysis.
• Training Manuals and Media: Training materials, observational manuals, pacing audio files, and standardized criterion video assessment guidelines are archived and accessible through Dr. McKenzie’s observational research repositories, institutional health networks, and published research protocols.

References

Below are core peer-reviewed academic references detailing the development, validation, and empirical implementation of the BEACHES direct observation instrument:

• Eboh, L. O., & Boye, T. E. (2005). Physical activity behaviours of female pupils and possible influences of urban environments. Pakistan Journal of Nutrition, 4(6), 361–365. https://doi.org/10.3923/pjn.2005.361.365
• Elder, J. P., Broyles, S. L., McKenzie, T. L., Sallis, J. F., Berry, C. C., Davis, T. B., Hoy, P. L., & Nader, P. R. (1998). Direct home observations of the prompting of physical activity in sedentary and active Mexican- and Anglo-American children. Journal of Developmental and Behavioral Pediatrics, 19(1), 26–30. https://doi.org/10.1097/00004703-199802000-00005
• McKenzie, T. L., Baquero, B., Crespo, N., Arredondo, E., Campbell, N., & Elder, J. P. (2008). Environmental correlates of physical activity in Mexican-American children at home. Journal of Physical Activity and Health, 5(4), 579–591. https://doi.org/10.1123/jpah.5.4.579
• McKenzie, T. L., Sallis, J. F., Elder, J. P., Broyles, S. L., Berry, C. C., Hoy, P. L., Nader, P. R., & Zive, M. (1997). Physical activity levels and prompts in young children at school recess: A two-year study of a bi-ethnic sample. Research Quarterly for Exercise and Sport, 68(3), 195–202. https://doi.org/10.1080/02701367.1997.10607998
• McKenzie, T. L., Sallis, J. F., Nader, P. R., Broyles, S. L., & Nelson, J. A. (1992). Anglo- and Mexican-American preschoolers at home and at recess: Activity patterns and environmental influences. Journal of Behavioral and Developmental Pediatrics, 13(3), 173–180. https://doi.org/10.1097/00004703-199206000-00005
• McKenzie, T. L., Sallis, J. F., Patterson, T. L., Elder, J. P., Berry, C. C., Rupp, J. W., Atkins, C. J., Buono, M. J., & Nader, P. R. (1991). BEACHES: An observational system for assessing children’s eating and physical activity behaviors and associated events. Journal of Applied Behavior Analysis, 24(1), 141–151. https://doi.org/10.1901/jaba.1991.24-141
• Sallis, J. F., Nader, P. R., Broyles, S. L., Elder, J. P., Berry, C. C., McKenzie, T. L., & Nelson, J. A. (1993). Correlates of physical activity at home in Mexican-American and Anglo-American children. Health Psychology, 12(5), 390–398. https://doi.org/10.1037/0278-6133.12.5.390

Items of the Scale