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 highly structured, direct observation instrument designed to simultaneously assess children’s health behaviors, including physical activity, sedentary behaviors, and eating habits, along with associated environmental characteristics and events. Developed within the framework of behavioral analysis, the system allows researchers to capture objective data on the social and physical variables that influence a child’s actions in naturalistic settings, such as homes and schools. The physical activity codes utilized in BEACHES have been validated against objective measures like heart rate monitors and accelerometers, ensuring high data integrity.

Keywords

Child health, Physical activity, Sedentary behavior, Eating behavior, Direct observation, Time sampling, Environmental correlates, Behavioral analysis, SOFIT.

Authors

Thomas L. McKenzie, Ph.D., J. F. Sallis, P. R. Nader, J. P. Elder, T. L. Patterson, C. C. Berry, J. W. Rupp, C. J. Atkins, M. J. Buono.

Purpose

The central purpose of the BEACHES system is to obtain objective, detailed data regarding children’s behaviors in their home or school environments, focusing specifically on physical activity and sedentary behaviors. It is designed not only to quantify the intensity and duration of these behaviors but also to identify the selected environmental variables (both social and physical) that may serve as prompts, antecedents, or consequences influencing these events.

This objective approach contrasts with self-report measures and is highly valuable for evaluating intervention programs aimed at modifying lifestyle behaviors in children. The system is flexible and has been adapted for various research contexts, such as the AVENTURAS study, which utilized a modified version focusing on seven key dimensions.

Construct

BEACHES primarily measures observable constructs related to health behavior and environmental context, grounded firmly in the framework of behavioral analysis. The core constructs assessed include the frequency and intensity of activity (Activity Level), the immediate physical context (Location), the social context (People There), and the motivational factors (Behavior Motivated and Motivator).

By using time sampling, the system captures dynamic interactions, allowing researchers to determine how specific environmental and social prompts (e.g., an adult encouraging movement or media viewing) directly influence the child’s subsequent actions, thereby quantifying the environmental correlates of energy expenditure and eating.

Validity

The validity of the BEACHES system, specifically concerning its physical activity codes, has been empirically established through concurrent validation methods. Researchers compared the direct observational codes of activity intensity with data simultaneously collected using physiological and objective measures, namely heart rate monitors and accelerometers. This rigorous validation process confirms that the behavioral codes accurately reflect the child’s actual level of energy expenditure, ensuring the instrument provides robust, objective data on movement behaviors.

Reliability

Reliability in the BEACHES system is centered on achieving and maintaining high inter-observer agreement (IOA) to mitigate observer error and drift. Assessors undergo extensive training, typically lasting around 16 hours, where they must first memorize the operational definitions of all behavioral dimensions and subcategories.

Training involves using videotaped examples and role-playing, culminating in live observations. An assessor is certified only after achieving an IOA score of 80% or greater on two criterion videotapes and two consecutive live field observations, calculated using interval-by-interval correspondence. To ensure data quality throughout long-term studies, observers are reassessed every three months, and those falling below the 80% criterion are immediately retrained.

Factor Analysis

BEACHES is an objective observation protocol designed for coding behaviors and environmental events, not a latent variable psychometric instrument. Therefore, traditional factor analysis used to identify underlying psychological factors is not applicable to the scale structure. Instead, the fidelity of the instrument rests on the clarity of its operational definitions and the demonstrated reliability of the observation intervals and categories, which are discrete and theory-driven (based on behavioral principles).

Instrument

Test Type: Direct Observation System (Behavioral Coding)

Format: Structured observation manual utilizing time sampling (15-second observation/15-second recording intervals), typically recorded manually on prepared forms or digitally via PDA/IPOD-paced prompts.

Language Available: English (Original development context, though implementation requires trained observers who may use other languages for interaction, the coding definitions remain English-based).

Population Group: Children and preschoolers.

Age Group: Typically used with young children (preschoolers and school-age children).

Population Details: Has been successfully applied in diverse ethnic populations, including Anglo-American and Mexican-American children, in home, school, and community settings.

Test Methodology: Trained assessors observe the target child for 60-minute sessions (two 30-minute segments) during specific times (e.g., after school until the evening meal). Data collection uses a combination of momentary time sampling (for Activity Level, Location, People There) and partial-interval time sampling (for Behavior Motivated, Motivator, Views Media, and Eats).

Keywords

Child observation, Health behavior, Environmental correlates, Time sampling, Accelerometry, Home environment, Behavioral assessment, SOFIT codes.

Authors

Author ORCID Identifier: Not provided in source material.

Affiliation Email addresses: Not provided in source material.

Correspondence Address: Not provided in source material. (Primary author: Thomas L. McKenzie, Ph.D.)

Permissions & Fee and Test Year

The original Behaviors of Eating and Activity for Children’s Health: Evaluation System (BEACHES) was first detailed and published in 1991 (McKenzie et al., Reference 1). The instrument is widely used in academic and public health research. Specific licensing fees or permissions required for commercial use are not detailed in the source material; researchers should contact the primary author or the institution where the research was conducted for permission to use the observational system.

Reference’s

1. 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, 141-151.
2. 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 Journal of Behavioral and Developmental Pediatrics, 13, 173-180.
3. 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, 390-398.
4. McKenzie, T. L., Sallis, J. F., Elder, J. P., Broyles, S. L., Berry, C. C., Hoy, P. L., Nader, P. R., Zive, M, & Broyles, S. L.. (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, 195-202.
5. 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, 26-30.
6. 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, 579-591.
7. Eboh, L. O. & Boye, T. E. (2005). Physical activity behaviours of female pupils and possible influences of urban environments on Pakistan Journal of Nutrition, 4(6), 361-365.

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

IMPORTANT: The following scale items must be preserved in their original language and must not be changed in any way.

The following Table briefly describes BEACHES dimensions to be employed in AVENTURAS:

Sample recording from (abbreviated). With 30 intervals per page=4 pages for 60 minutes.

Observation Procedures

Home includes both inside and outside the home or apartment complex. Visit homes only on days a parent/guardian confirms the target child will be there for at least 2 hours.

BEACHES observations will not be made when the child is away from home (e.g., at a friend’s house or in a park area that is not part of the apartment complex).

Observers focus on the target child for a 15-second observation interval and then have 15 seconds to enter data codes onto prepared forms. Voice prompts on an IPOD or MP3 pace the alternating observation and recording periods. Each observe-record cycle requires 30 seconds; thus, a 90-minute period would produce 180 observation intervals.

Three dimensions (Child Activity, Location, and People There) are scored using momentary time-sampling methods (i.e., codes are entered to describe events related to these three categories as they occur at the end of the “observe” interval). The other four dimensions (Activity Motivation, Activity Interactor, Views Media, and Eats) are scored using partial-interval time sampling (i.e., events are coded if they occur at any time during the 15-second “observe” interval).

Observer Training and Calibration

Assessors memorize operational definitions of the behavior dimensions and their subcategories first and then learn the general procedures for recording data. During assessor training videotaped examples and role-playing are used to demonstrate each category. This is followed by live observations at homes. Training for each observer continues until she exceeds an inter-observer agreement score of 80% on two different criterion videotapes, and 80% on two consecutive live observations (using interval-by-interval correspondence, with agreements divided by agreements plus agreements multiplied by 100). Observers are also trained how to interact in order to reduce reactivity. The average training program takes about 16 hours.

Throughout the data collection period additional review and training sessions approximately one hour in length will be conducted at least every semester. To ensure maintenance of data quality and to guard against observer drift, observers should be reassessed once every three months through the independent coding of a videotape that has previously been coded by the designers of BEACHES. Any observer scoring below criteria level (80% agreement) should be retrained until mastery is once again achieved.

Each observer should complete at least five reliability assessments in the field during each data collection period.

The assessor, on a rotational basis throughout the entire period, observes the target child for 15 seconds and then takes up to 15 seconds to enter codes for seven variables. A combination of momentary and partial time sampling is used.

CATEGORIES 1.0, 2.0, 3.0,

Use momentary time sampling. Code these three variables (ACTIVITY LEVEL, WHERE, and PEOPLE THERE) only as they are occurring at the end of the observe interval (i.e., on the “record” signal).

CATEGORIES 4.0 to 7.0

Use partial interval time sampling. That is, code relevant events if they occur at any time during the observe interval.

1.0 ACTIVITY LEVEL

Code the activity level that the child was engaging in at the “record” signal. BEACHES activity codes are the same as those used in SOFIT.

• √ Code 1-4 (Lying Down, Sitting, Standing, Walking) to describe the body position of the child only if he/she in not expending more energy than that required for an ordinary walk.
• √ Code 5 (Very active) for any activity in which the child was expending more energy than he/she would during ordinary walking, irrespective of body position. For example, code 5 (Very active) if the child is wrestling with a peer (even though he is lying on his back), standing on his hands, or pedaling a bicycle vigorously (even though sitting).
• √ When the child is in transition from one category to another, enter the code for the higher category. For example, code ‘2’ (sitting) if at the “record” signal the child is partially lying down and partially sitting up; code ‘3’ (standing) if he/she is getting up from either sitting or lying down.

Sample activity codes:

kneeling (weight on knees only) = standing kneeling (weight on knees and buttocks) = sitting inactive “on all fours” = standing

seated swinging (arms producing no momentum)= sitting seated swinging (arms producing momentum)= very active obviously struggling to push or pull objects = very active

2.0 WHERE (LOCATION)

Code where the child was located at the “record” signal.

• √ -I (Inside). Child is inside the home (includes hallways)
• √ -O (Outside). Child is outside the home (in yard; on sidewalk or street) or in a home location specifically designed for physical activity (e.g., pool or activity room at apartment complex).

3.0 PEOPLE THERE

Code all persons in the child’s immediate environment at the record signal.

• √ -N (None). No one else is present
• √ -P (Parent/Guardian). At least one parent or guardian is present.
• √ -S (Sibling) is present. At least one brother or sister is present
• √ -OC (Other child). A child other than the child’s brothers or sisters is present.
• √ -OA (Other Adult). An adult other than the child’s parent or guardian is present.

For this category, persons are considered to be part of the environment only if they are proximally located to the target child (i.e., typically within 25 feet) and within the designated boundaries of the environment (e.g., in the same room). Observers should determine the boundaries of the environment before beginning data collection. Look for natural breaks in play, work, and eating areas that are separated by walls or shrubbery.

4.0 BEHAVIOR MOTIVATED

Code events related to motivating/influencing the target child to directly engage in either physical activity or sedentary behavior. These events may be antecedents (e.g., prompts) or consequences (e.g., rewards or punishers).

• √ -N (None). No prompts or consequences were provided for the target child to engage in either physical activity or sedentary behavior during the interval.
• √ -PA (Physical Activity). A person provided a prompt or consequence to directly motivate the target child to initiate or continue to engage in physical activity (i.e., at activity levels 4 or 5).
• √ -SB (Sedentary Behavior). A person provided a prompt or consequence to directly motivate the target child to initiate or continue to engage in sedentary behavior (i.e., level 1, 2, or 3).

Physical activity refers to those activity levels that would be coded as Activity Levels 4 and 5. References to (i.e., prompts and consequences related to) all locomotor activities (e.g., walking, running, hopping, skipping, galloping, chasing, fleeing, dodging, and crawling), as well as manipulative (e.g., throwing, catching, kicking, punting, dribbling, volleying, and striking), and non-manipulative activities (e.g., balancing, rolling, twisting, hanging, jumping) that involve gross motor movement are coded as physical activities.

Examples with including interactor coded–Dimension 5.0):

• Mom says: “Go outside and play.” Code as “Adult; motivates physical activity.” Boy tosses study child a ball. Code as Child; motivates physical activity.”
• Study child is playing tag with peers. Code as Child; motivates physical activity.”
• Study child is playing chess or video games with sibling. Code as Child, motivates sedentary behavior.”
• Study child sits with friend on couch, but do not interact. Code (“N, N”) to indicate there was no behavior motivated, and thus no interactor.
• Sibling says: “Let’s stop jumping rope now.” Code as Child, motivates sedentary behavior.”
• Grandfather says, “Stop all that running around.” Code as “Adult; motivates sedentary behavior.”
• During an observe interval TC is engaged reciprocally in active play with a girl (e.g., during wrestling, tag).
• Code as Child: motives physical activity.”
• Boy on monkey bars says to TC, “See how much fun playing on this bar is.” Code as Child, motivates physical activity.” (i.e., the boy provides imitative prompt)
• When the study child is a participant in a group, code prompts and consequences provided to that group. For example if T.C. is a member of the Striders and hears: “All the children in the Striders Group did such a good job sprinting to the finish,” code as “Adult, motivates physical activity.”
• Code all prompts that relate specifically to doing gross body activities. For example, mom says “Go to the refrigerator and get the milk.” Code as “Adult, motivates physical activity.” (i.e., walking). However, do not code statements such as “Pass the butter.” or “Get your feet off the table and sit up straight.” as prompts for either physical activity or sedentary behavior.
• Motivators (i.e., prompts and consequences) are coded only if likely (intended) to have an immediate (within that day) effect (are contingent) on physical activity or sedentary behavior.
• Stated motivators (i.e., verbal) must be direct and refer to immediate specific and concrete events, and not refer to remote and vague concepts such as references to becoming fat, fit, or sick.
• In cases when there are multiple prompts or consequences during an interval, record only the first prompt or first consequence AND its interactor.
• Record verbal consequences that are made regarding and heard by the study child but spoken to others. For example, code when a study child hears his mom say to dad, “Look at that T.C., I’m so proud of the way he’s exercising” as “Adult, motivates physical activity.”
• Do not code verbal prompts if you don’t hear them clearly. (Listen carefully and keep up with study child.) Do not code requests to be physically active or engage in sedentary behavior by the target child.

5.0 MOTIVATOR

Code a motivator only if someone during the 15-second observe interval participated in a physical or verbal exchange that was related to directly motivating/influencing physical activity or sedentary behavior. The person coded is associated with the Activity Motivated in Dimension 4.0.

Physical activity behaviors refer to those that would be coded in Category 1 as a “4” (Walking) or “5” (Very Active).

• √ -N (None). No interactions were related to motivating physical activity or sedentary behavior during the interval.
• √ -C (Child). The motivator of the physical activity or sedentary behavior was a child.
• √ -A (Adult). The motivator of the physical activity or sedentary behavior was an adult (e.g., parent, other adult).

6.0 VIEWS MEDIA

• √ -Y (Views). Code “Y” if a television, VCR, DVD, computer or a video or computer game in the child’s direct environment is turned on and the child focuses on it for at least three seconds during the interval
• √ Do not consider whether the viewing was for play or educational purposes.

7.0 EATS (EATING BEHAVIOR)

• Code whether or not the child ingested food during the observe interval.
• √ Code ingests food (Y) only if the child puts food into his/her mouth at any time during the “observe” interval.
• √ Do not code for food that is already in the child’s mouth or for food that the child removes from his/her mouth and re-inserts it
• √ Medicine and gum are considered food, but water is not.