Table 2 Description of development and testing methods for parental feeding practice instruments

Jensen (1983) Family Routines Inventory [24, 25]

4

3: interviews with families, literature review

mothers ranking of most important routines

Test-retest: r = 0.79

Construct Validity: total score on new survey was significantly correlated with the Family Environment Scale’s cohesion (rho = 0.35), organization (rho = 0.36), control (rho = 0.20, and conflict (rho = -0.18) scales.

Stanek (1990) Eating Environment [26]

2

2: expert opinion

expert review, pilot of survey

Construct Validity: child helps prepare food, child allowed to decide type of food eaten, use of small portions when introducing new foods, use discussion to persuade child to eat, leave child alone if refusing to eat, praise child for eating healthy were all associated with intake of foods from basic food groups (r = 0.18, p < .01).

Seagren (1991) Parents’ Behavior and Attitudes Toward their Children’s Food Intake [27]

4

3: expert opinion, observation and interviews in WIC clinics, literature review

expert review

Construct Validity: parents of overweight children were significantly less likely to report controlling the type of foods allowed for snacks, allowing sweets only after a healthy meal, encouraging child to eat all food on plate, and encouraging child to eat as much as they would like.

Sherman (1992) Maternal Feeding Practices Questionnaire [28]

3

2: pulled from existing surveys

Construct Validity: Pushier feeding practices was not significantly correlated with child weight.

Davies (1993) About Your Child’s Eating [29, 30]

3

2: expert opinion

factor analysis

0.38-0.73

Construct Validity: AYCE factors correlated significantly and in expected directions with the Family Environment Scale factors (r = -0.9-0.39, p < .05 for all).

Structural Validity: Final CFA model good fit, NFI = 0.87, NNFI = 0.91, R-CFI = 0.93, RMSR = 0.04, and SB-χ²/df = 1.75

Crist (1994) Behavioral Pediatrics Feeding Assessment Scale [31]

3

2: pulled from existing surveys

None

Test-retest: r = 0.83

Sallis (1995) Study of Children’s Activity and Nutrition [32]

3

1

Koivisto (1996) Mealtime Practices [33]

3

2: pulled from existing surveys

factor analysis

>0.30

Humphry (1997) Feeding Stories [34]

4

3: interviews with parents, literature review

factor analysis

not reported

Test-retest: r = 0.68-0.90

De Bourdeaudhuij (1998) Interactions Around Food [35, 36]

4

2: pulled from existing surveys, open ended questions

Inter-rater: Pearson r = 0.02-0.49

Construct Validity: regression models showed that negative strategies was a significant predictor of child’s healthy food score (β = -0.17) and veg intake (β = -0.19); and obligation rules was a significant predictor of soda intake (β = -0.35).

Golan (1998) Family Eating and Activity Habits Questionnaire [37, 38]

3

2: expert opinion, literature review

expert review, pilot of survey, factor analysis

not reported

Test-retest: r = 0.78 -0.90 Inter-rater: r = 0.81-0.94

Construct Validity: T-tests comparing scores from obese and normal-weight children showed that obese children have significantly higher scores on all scales and for total score (F(1,37) = 11.5).

Hupkens (1998) Food Rules [39]

3

2: qualitative study

pilot of survey

Fisher (1999) Parental Restriction [40]

4

1

Construct Validity: Maternal use of restriction was significantly correlated with child selection of the restricted food (r = 0.41) and child weight for height (r = 0.42).

Carper (2000) Kids' Version of the Child Feeding Questionnaire [41]

4

2: pulled from existing survey

Inter-rater: pressure to eat was the only parent- reported variable that significantly predicted daughters’ perception (OR = 1.5)

Construct Validity: girls' perceived pressure to eat was significantly associated with dietary restraint (OR = 3.0), emotional disinhibition (OR = 3.2), and external disinhibition (OR = 3.0), and perception of restriction was significantly associated with external disinhibition (OR = 0.4).

Cullen (2000) Parent Food-Socialization Practices [43]

4

2: pulled from existing surveys, focus groups with parents

factor analysis

0.41-0.89

Test-retest: r = 0.61-0.89

Construct Validity: Dinner FJV preparation was significantly correlated with child juice intake (r = -0.35).

Neumark-Sztainer (2000) Project EAT [46–48]

4

3: pulled from existing surveys, expert opinion, focus groups with youth

poor test-retest or internal consistency

Test-retest: r = 0.54-0.70

Construct Validity: regression model showed that social support for healthy eating and family meal patterns were significant predictors of child F&V intake, but only had indirect effect through home F&V availability.

Structural Validity: 2003: Final CFA model had good fit, factor loadings were 0.37-0.82, χ² (347) = 3099, NFI = 0.99, RMSEA CI 0.043, 0.046.

Ross (2000) Family Unpredictability Scale [49]

4

3: expert opinion

expert review, factor analysis, pilot with parents

0.47-0.85

Construct Validity: meals was significantly correlated with other measures of family functioning (r = 0.18-0.31). CFA: Final higher order model.

Structural Validity: final CFA higher order model had good fit, (df) χ² = 87 (102.9), GFI = 0.90, CFI = 0.96, AGFI = 0.86, RMSEA = 0.04, AIC = 168.9, AIC-S = 240, AIC-I = 630.6.

Baughcum (2001) Preschooler Feeding Questionnaire [50]

4

2: pulled from existing surveys, expert opinion, focus groups with dieticians and mothers, literature review

factor analysis

0.49-0.82

Construct Validity: Scores on relevant factors were not significantly different between parents of normal vs. overweight children.

Birch (2001) Child Feeding Questionnaire [52–54]

4

2: pulled from existing survey, findings from previous research

factor analysis

0.37-0.95

Construct Validity: In sample 1, pressure to eat (r = -0.26) and restriction (r = 0.13) were significantly correlated with child weight. In sample 2, only responsibility (r = 0.20) was significantly correlated. In earlier study, controlling practices were significantly correlated with child’s ability to compensate for caloric density (r = 0.65).

Structural Validity: Final CFA model in sample 1 had good fit, χ² (229) = 419, CFI = 0.95, NNFI = 0.94, RMSEA = 0.04; final model in sample 2 confirmed with minor modifications, χ² (227) = 309, CFI = 0.92, NNFI = 0.91, RMSEA = 0.05; final model in sample 3 confirmed after 3 items removed, χ² (166) = 232, CFI = 0.91, NNFI = 0.89, RMSEA = 0.05.

Criterion Validity: Mothers’ reported practices were not correlated with observed mealtime behaviors. Fathers’ reported pressure to eat was significantly correlated with observed use of pressure (0.36), prompting (0.65), and use of incentives (0.44); and reported restriction was significantly correlated with observed use of pressure (0.37) and use of incentives (0.47).

Cullen (2001) Family and Peer Influences on FJV Intake [61]

4

2: pulled from existing surveys, focus groups

factor analysis

0.43-0.85

Test-retest: r = 0.19-0.59

Construct Validity: Parent FJV modeling was significantly correlated with child intake of fruit (r = 0.18), juice (r = 0.14), total FJV (r = 0.20); and parent control was significantly correlated with child juice intake (r = 0.17).

Tibbs (2001) Parental Dietary Modeling Scale [62, 63]

4

2: focus groups with parents, literature review

Construct Validity: Tibbs found that modeling was significantly associated with eating patterns (r = 0.48), low fat eating (r = -0.30), and F&V intake (r = 0.18). Moens found that parental modeling did not differ significantly between normal and overweight children,and parental modeling did not contribute to the prediction model snack intake.

Tiggemann (2002) Control Over Child Feeding [64]

4

1

factor analysis

0.53-0.83

Construct Validity: monitoring was significantly correlated with child BMI (r = 0.30) and BMI % (r = 0.33).

Wardle (2002) Parental Feeding Style Questionnaire [65]

4

3: pulled from existing surveys, interviews with mothers, literature review

cognitive interviews, pilot

Test-retest: r = 0.76-0.83

Construct Validity: Prompting/encouragement to eat was the only scale significantly correlated with child BMI (r = 0.19), and only significant for first-born twins.

Bourcier (2003) Eating for a Healthy Life – Strategies to Influence Eating Behavior [68]

3

2: pulled from existing surveys

factor analysis

not provided

Construct Validity: Reliance on self was a significant factor in model predicting child fat intake (B = -1.35, SE = 0.07); and pressure was a significant factor in the model predicting F&V intake (B = 1.44, SE = 0.04).

Cullen (2004) GEMS - Diet-Related Psychosocial Questionnaire [69]

3

2: pulled from existing surveys

factor analysis

0.37-0.73

Test-retest: ICC = 0.66-0.69

Construct Validity: low-fat food preparation practices was significantly associated with lower percent energy from fat (r = 0.23); high-fat food preparation practices was significantly associated with higher percent energy from fat (r = 0.24).

Melgar-Quinonez (2004) Child Feeding Strategies [70, 71]

2

2: focus groups with parents

factor analysis

0.67-0.76

Construct Validity: Multivariate analysis did not find any of the scales to be associated with child overweight; however, child takes food from refrigerator or panty between meals was significantly associated with obesity (OR = 0.32)

Vereecken (2004) Food Parenting Practices [72, 73]

4

2: discussions with parents, literature review

Construct Validity: permissiveness was significantly correlated with child intake of veg (r = -0.16), soda (r = 0.59), and sweets (r = 0.23). Pressure was significantly correlated with intake of veg (r = 0.15). Material reward was significantly correlated with intake of sweets (r = 0.19). Verbal praise was significantly correlated with intake of fruit (r = 0.16), veg (r = 0.20), and soda (r = -0.14). Negotiation was significantly correlated with intake of veg (r = 0.19). Encouragement was significantly correlated with intake of fruit (r = 0.22). Catering on demand was significantly correlated with intake of veg (r = -0.14), soda (r = 0.15), and sweets (r = 0.16). Full regression model found that permissiveness was a significant predictor of soda intake (OR = -8.81), material reward was a significant predictor of sweets intake (OR = 1.54), and praise was a significant predictor of veg intake (OR = 1.38).

De Bourdeaudhuij (2005) Pro Children Project [74]

4

3: expert opinion, focus groups with children, interviews with parents and staff, literature review

cognitive interviews, poor test-retest

Test-retest: r = 0.50-0.73

Construct Validity: active parent encouragement was significantly correlated with child intake of fruit (r = 0.17) and veg (r = 0.24); demand family rule was significantly correlated with child intake of fruit (r = 0.22) and veg (0.15); and allow family rule was significantly correlated with child intake of veg (r = 0.17).

Horodynski (2005) Child-Parent Mealtime Behavior Questionnaire [75]

3

2: pulled from existing survey

factor analysis

not reported

Hughes (2005) Caregiver’s Feeding Style Questionnaire [6, 76]

4

2: pulled from existing surveys, cognitive interviews, videotaped observations of mealtimes, literature review

factor analysis, low variability

0.30-0.65

Test-retest: ICC = 0.82-0.85

Construct Validity: parents with indulgent feeding style were more likely to have overweight children compared to authoritarian parents (F (3, 227) = 2.19, p < 0.04). Also noted significant main effects for feeding styles with the CFQ (F (9, 518) = 3.17) and the Parenting Dimensions Inventory (F (27, 602) = 2.26)

Tripodi (2005) Family Dietary Habits (part of the Italian National Institute of Nutrition) [78]

1

2: pulled from existing surveys

Construct Validity: parent behavior when child refuses to eat was significantly associated with child BMI (β = 0.86).

Vereecken (2005) Social and Environmental Influences on FJV Consumption [79]

4

2: pulled from existing surveys, open ended survey questions, literature review

team reviewed items from existing tools and reduced

Test-retest: ICC = 0.44-.071

Construct Validity: perceived parent behavior was significantly correlated with child intake of fruit (r = 0.30) and veg (r = 0.45); and permissiveness was significantly correlated with veg intake (0.15).

Arredondo (2006) Parenting Strategies for Eating and Activity Scale [80, 81]

3

2: Pulled from existing surveys, focus groups with mothers

factor analysis

not provided

Construct Validity: Arredondo found that monitoring (β = 0.45), reinforcement (β = 0.32) and discipline (β = 0.20) were significantly associated with healthy eating; and monitoring (β = -0.17), reinforcement (β = -0.08) and control (β = 0.10) were significantly related to unhealthy eating. Noted some interactions with child gender. Larios found that control was significantly associated with child BMI (r = -0.21, p < 0.01). Also noted significant associations between PEAS scales and Birch’s CFQ scales.

Structural Validity: In Arredondo, CFA model fit was good, χ² (279) = 2.79, RMSEA = 0.06. In Larios, CFA final model fit was good, χ² (282) = 1030.81, CFI = 0.89, IFI = 0.90, RMSEA = 0.06

Ogden (2006) Overt and Covert Control [82–84]

4

2: discussions with mothers, literature review

factor analysis

0.54-0.81

Construct Validity. overt control was significantly correlated with CFQ’s restriction (r = 0.27), pressure to eat (r = 0.46), and monitoring (r = 0.39); and covert control was also significantly associated with the 3 CFQ subscales (r = 0.42, 0.26, and 0.42). Regression models also showed that covert control predicted of unhealthy snack food and F&V intake, and overt control predicted F&V intake. Neither covert or overt control helped predict child BMI.

de Moor (2007) Management Techniques of Feeding Problems [85]

3

2: pulled from existing surveys

factor analysis

0.33-0.84

Construct Validity: positive behavioral support was significantly correlated with child pickiness (r = 0.47) and disturbing mealtime behavior (r = 0.46); as was negative behavioral support (r = 0.38 and 0.47) and general management technique (r = 0.17 and 0.28).

Gray (2007) Parental Attitudes around Feeding [86]

4

2: pulled from existing surveys, expert opinion, and literature review

expert review

Not reported

Construct Validity: parents of overweight or at risk for overweight children were significantly more likely to disagree with statement about encouraging the child to eat more.

Musher-Eisenman (2007) Comprehensive Feeding Practices Questionnaire [87]

4

3: pulled from existing surveys, open ended survey items, systematic review

factor analysis, eliminated items that were confusing or had no variability

0.31-0.95

Structural Validity: final CFA model fit was good, χ² (1061) = 1580, CFI = 0.98, RMSEA = 0.057

Reinaerts (2007) Social Influence on F&V Consumption [89]

3

2: interviews with children and parents, literature review

factor analysis

0.66-0.91

Construct Validity: regression models showed that parent modeling of F&V intake was significant predictor child F&V intake (β = 0.0-0.34).

Stanton (2007) Diet-Specific Social Support [90]

3

2: pulled from existing surveys

factor analysis

0.52-0.81

Construct Validity: family support was a significant predictor of fiber intake (β = 0.23)

Vue (2007) Individual and Environmental Influences on Calcium Intake [91]

3

3: pulled from existing surveys, focus groups with children

factor analysis

0.41-0.71

Construct Validity: independence was significantly correlated with child intake of cheese (r = 0.24), parental expectations was significantly correlated with intake of soy milk (r = 0.21), parental modeling was significantly correlated with intake of soda (r = -0.24), OJ (r = -0.23), and dark green veg (r = -0.29), family limitations was significantly correlated with intake of soda (r = -0.19) and cheese (r = 0.31).

Bryant (2008) Healthy Home Survey [92]

3

2: pulled from existing surveys, expert opinion, literature review

expert review

Test-retest: percent agreement = 42.2-97.8, Kappa = 0.36-0.88, and ICC = 0.32-0.93

Criterion Validity: (only available for food environment items): percent agreement = 57.7-92.3, Kappa = 0.07-0.57.

Burgess-Champoux (2008) Determinants of Whole Grain Intake [93]

4

3: pulled from existing surveys, focus groups with parents, literature review

pilot of survey, factor analysis

0.46-0.89

Test-retest: Not reported for relevant scales

Byrd-Bredbenner (2008) Food Decision Influencer [94]

4

2

Construct Validity: cluster analysis identified 4 clusters: (1) happy, healthy food involved mothers, (2) working, convenience driven mothers, (3) healthy, free of food price, taste, convenience, and advertising effects mothers, and (4) stressed, emotional eating, time-conscious mothers. Cluster 1 had significantly lower mother and child BMIs compared to other clusters.

Faith (2008) Feeding Demands Questionnaire [95]

3

2: expert opinion

factor analysis

0.78-0.89

Construct Validity: across 2 samples, total FEEDS score was significantly associated with CFQ’s monitoring (r = 0.30-0.36) and pressure to eat (r = 0.41-0.53); anger/frustration subscale was significantly associated w CFQ’s pressure to eat (r = 0.32-0.47); food amount demandingness was significantly associated with CFQ’s monitoring (r = 0.29-0.45), restriction (r = 0.24-0.26), and pressure to eat (r = 0.38-0.46); and food type demandingness was significantly associated with CFQ’s monitoring (r = 0.36-0.43). None of the scales were consistently associated with child BMI z score.

Fulkerson (2008) Family Meals [96]

4

2: pulled from existing surveys

Gattshall (2008) Home Environment Survey [97]

3

2: pulled from existing surveys

factor analysis, item performance (low variability, extreme means or low correlation with scale)

not provided

Test-retest: ICC = 0.80-0.82 Inter-rater: ICC = 0.24-0.54;

Construct Validity: parental role modeling was significantly correlated with child’s intake of fruit (r = 0.21) and veg (r = 0.14); and parental policies was significantly correlated with child intake of fruit (0.28) and veg (0.36).

Haerens (2008) Home Environment Related to Eating [98]

2

2: pulled from existing surveys

Test-retest: ICC = 0.88-0.89 Inter-rater: r = 0.54-0.66

Construct Validity: food rules significantly contributed to prediction model for boys’ fat intake (β = 0.14) and girls’ fruit intake (β = -0.16); and TV viewing contributed to boys’ intake of soft drinks (β = 0.14) and fruit (β = -0.22) and girls’ intake of fat (β = 0.15) and fruit (β = -0.10).

Haire-Joshu (2008) High 5 for Kids [99]

2

2: pulled from existing surveys, focus groups

Test-retest: ICC = 0.50-0.66

Construct Validity: changes in parent modeling and use of non-coercive feeding did not predict changes in child F&V intake.

Kroller (2008) Parental Feeding Strategies [100]

3

2: pulled from existing surveys, expert opinion, interviews with mothers

Test-retest: r = 0.41-0.78

Construct Validity: regression model showed that pressure (β = 0.12) was a significant predictor of child intake of problematic foods; and child control (β = 0.24) and rewarding (β = -0.26) were significant predictors of child F&V intake.

Spurrier (2008) Physical and Nutritional Home Environment Inventory [101]

3

2: pulled from existing surveys

Construct Validity: portion size served, foods eaten in front of TV, acceptance of wasted food, reminding of child to ’eat up’, offering food as reward, and restriction of juice/high-fat and high-sugar foods/second helpings were significantly associated with child F&V intake; frequency of family meals, meals in front of TV, use of food to reward good behavior, and restriction of juice/carbonated beverage were significantly associated with child’s sweetened beverage intake; and use of food ’treats’ as reward for eating main meal, restriction of juice/high-fat and high-sugar foods, carbonated beverages, and snack/meals in front of TV were associated with child’s intake of non-core foods.

Hendy (2009) Parent Mealtime Action Scale [102]

4

2: pulled from existing surveys, literature review

factor analysis

0.42-0.87

Test-retest: r = 0.51-0.75 Inter-rater: r = 0.59-0.78

Construct Validity: positive persuasion (β = 0.07), daily F&V availability (β = 0.32), and special meals (β = -0.07) were significant predictors of child F&V intake; positive persuasion (β = 0.08), snack modeling (β = 0.17), fact reduction (β = -0.08), and many food choices (β = 0.08) were significant predictors of child snack intake; and positive persuasion (β = -0.08), insistence on eating (β = -0.12), snack modeling (β = 0.09), and fat reduction (β = 0.12) were significant predictors of child BMI%.

Structural Validity: factor loading from CFAs in 3 samples ranged from 0.21-0.85, but no model fit indices were reported.

Joyce (2009) Parent Feeding Dimensions Questionnaire [103]

3

1

factor analysis referenced, but never published

not reported

Construct Validity: none of the feeding dimension were significantly correlated with child BMI; however, coerciveness was significantly correlated with child disinhibited eating (r = 0.16).

Neumark-Sztainer (2009) Ready Set ACTION [104]

2

2: pulled from existing surveys

Pearson (2009) Parental Modeling and Support [105]

3

1

Construct Validity: parent modeling of breakfast was positively associated with F&V consumption in boys and girls (boys OR = 1.53, girls OR = 1.66).

Corsini (2010) Toddler Snack Food Feeding Questionnaire [106]

4

2: interviews with mothers

pilot with parents

0.21-0.82

Test-retest: ICC = 0.79-0.90

Construct Validity: rules and CFQ monitoring (0.40, 0.45), flexibility (ns, -0.32), and allow access (-0.21, -0.39) were significantly associated with CFQ monitoring; allow access was also significantly associated with CFQ restriction (0.28, ns). Also, rules was significantly associated with chip intake (-0.25); flexibility was significantly associated with intake of savory biscuits (0.20), sweet biscuits (0.18), chips (0.19), and high fat/sugar dairy (0.17); and allow access was significantly associated with intake of savory biscuits (0.38), sweet biscuits (0.42), cakes and pastries (0.28), chips (0.52), and high fat/sugar dairy (0.38). No significant associations with child BMI.

Dave (2010) Home Nutrition Questionnaire [107]

4

3: pulled from existing surveys, focus groups and interviews with mothers

cognitive interviews, factor analysis

0.58-0.87

Construct Validity: regression model showed that parental practices to promote F&V intake (β = 0.61) and role modeling (β = 0.34) were significant predictors of home F&V availability and accessibility.

MacFarlane (2010) Adolescent Perceptions of Parent Feeding Practices [108]

2

2: pulled from existing surveys

factor analysis

0.51-0.89

Construct Validity: significant score differences between parents who were concerned vs. not concerned with child weight were observed for negotiation (-0.17 vs. 0.06, p < 0.001) and pressure to eat disliked food items (-0.08 vs. 0.03, p = 0.05).

McCurdy (2010) Family Food Behavior Survey [109]

4

2: pulled from existing surveys

cognitive interviews, factor analysis

0.43-0.90

Test-retest: ICC > 0.65

Construct Validity: all scales were significantly correlated with at least one other scale. Child choice was significantly correlated with maternal control (-0.47) and organization (0.34); and maternal control was significantly associated with maternal presence (0.34). Mothers with overweight children also had higher scores on maternal control (t(23) = 2.06, p = 0.052), but only at time 1. Mothers of normal weight children had higher scores on maternal presence (t(19) = -2.85, p = 0.01), but only at time 2.

O’Connor (2010) Food Parenting Practices [110]

4

3: expert opinion, focus groups with parents

Construct Validity: practical methods was significantly correlated with child F&V intake (r = 0.08), and firm discipline was significantly associated with child BMI z-score (r = -0.14). Neither parent practice categories or clusters contributed significantly to the model of child F&V.

Tremblay (2010) Quebec Longitudinal Study of Development – Meal Interactions [111]

3

1

Construct Validity: in boys, meal conflict had a direct effect on child body weight (more conflicts, higher body weight) and healthy eating (more conflicts, healthier eating). In girls, meal conflict had a direct effect on healthy eating (more conflicts, healthier eating).

Structural validity: For boys, Nχ² = 2.48, CFI = 0.94, SRMR = 0.04, RMSEA = 0.05; and for girls, Nχ² = 1.68, CFI = 0.97, SRMR = 0.03, RMSEA = 0.03.

Zeinstra (2010) Parental Child-Feeding Strategies [112]

3

2: pulled from existing surveys

factor analysis, pilot with parents

0.49-0.82

Construct Validity: regression models showed that choice (β = 0.28), distraction (β = -0.13), negative atmosphere (β = -0.19), pressure (β = 0.21), and positive info (β = -0.13) were significant predictors of child vegetable intake; and choice (β = 0.17) and negative atmosphere and pressure (β = -0.12) were significant predictors of child fruit intake. Correlations were also observed between CFQ scales on this new survey.

Berlin (2011) Feeding Strategies [113]

3

2: expert opinion

factor analysis, agreement between project team members ratings of items’ potential fit with constructs

0.33-0.89

Construct Validity: r = -0.43-0.46, significant correlations were observed between: across 2 samples, mealtime structure was significantly correlated with meal schedule (0.38, 0.45), child control of intake (-0.16, 0.12), parent control of intake (ns, 0.11), and between meal grazing (-0.27, -0.37); meal schedule was significantly correlated with parent control (ns, 0.14), between meal grazing (-0.28, -0.30), and clean plate (ns, 0.12); child control of intake was significantly correlated with parent control (ns, -0.34), between meal grazing (0.18, 0.22), and clean plate (ns, -0.27); parent control of intake was significantly correlated with between meal grazing (ns, -0.19), and clean plate (0.46, 0.38); between meal grazing was significantly correlated with clean plate (ns, -0.15). All but encourages clean plate were significantly correlated with one or more scales from the About Your Child’s Eating Scale.

Structural Validity: For the community sample, SB χ = 980.43, df = 448, CFI = 0.91, RMSEA = 0.064 (90% confidence interval: 0.058-0.069).

Byrd-Bredbenner (2011) Social Cognitive Theory Concepts [114]

4

2: pulled from existing surveys, expert opinion, literature review

Construct Validity: mothers scoring in the lowest tertile for plans meals had significantly higher BMIs compared to those in the highest tertile (p = 0.0031, F = 3.531).

McIntosh (2011) Family Meal Rituals [115]

3

2: pulled from existing surveys

factor analysis

not reported

Construct Validity: logistic regression showed that father’s perception of the family dinner as an important family ritual was a significant predictor of use of fast-food restaurants (OR = 0.39).

Moreno (2011) Family Health Behavior [116]

2

2: pulled from existing surveys, expert opinion, responses to caregiver survey, literature review

expert review, factor analysis, item performance

0.43-0.75

Test-retest: ICC = 0.75-0.77

Construct Validity: binary logistic regression showed that for every point increase in the total score, there was a 3.9% decrease in likelihood of child being overweight or obese (OR = 0.92, p < 0.01). However, bivariate correlations did not show significant associations between relevant scales (parent behavior, mealtime routines) and child zBMI.

Murashima (2011) Parental Control Over Child Feeding [117, 118]

4

3: pulled from existing surveys, expert opinion

expert review, cognitive interviews, factor analysis

0.45-0.83

Test-retest: r = 0.45-0.85

Construct Validity: high control (-0.14) and high contingency (-0.13) were significantly associated with child BMI; child centered strategies (0.20), encouraging nutrient dense foods (0.26), and timing of meals (-0.12) were significantly associated with intake of nutrient dense foods; and encouraging nutrient-dense foods (-0.12) and discouraging energy dense foods (0.26) were significantly associated with intake of energy dense foods.

Structural Validity: χ² = 292, df = 179, p < 0.05, CFI = 0.927, RMSEA = 0.044 (after timing of meals was removed).

Stifter (2011) Baby’s Basic Needs [119]

3

1

Construct Validity: feeding to soothe was significantly correlated with pressuring (0.23) and indulgent (0.23) styles; and interaction of using food to soothe child and child negativity was a significant predictor of child BMI z-score (p = 0.012).

Anderson (2012) Meals in Our Household [120]

4

2: pulled from existing surveys, expert opinion, literature review

low item correlations were used to remove items from scales

Test-retest: r = 0.80-0.95

Construct Validity: across 2 samples, family meals was significantly correlated with problem behaviors (-0.51, -0.38), parental concern (-0.29, -0.52), food as a reward (-0.21, ns), and spousal stress (-0.35, -0.23); food as a reward was significantly correlated with problem behaviors (0.33, 0.52), parental concern (0.33, 0.46), spousal stress (0.31, 0.46), and child influence (0.24, 0.35); and child influence was significantly correlated with problem behaviors (0.31, 0.48), parental concern (0.36, 0.49), and spousal stress (0.47, 0.38).

Dave (2012) Parental Social Support [121]

4

2: pulled from existing surveys, expert opinion, literature review

factor analysis

0.42-0.99

Test-retest: r = 0.56-0.94

Construct Validity: instrumental support (0.25) and positive encouragement (0.15) were significantly associated with F&V availability; and instrumental support (0.45), positive encouragement (0.29), and reinforcement (0.19) were significantly associated with F&V accessibility.

Moore (2012) West Virginia Healthy Lifestyle Act Evaluation [122]

1

2: pulled from existing surveys

Construct Validity: logistic regression showed that parents who were concerned with their child’s weight were significantly more likely to report trying to change their family’s diet to make it healthier, put their child on a diet, and have their child skip meals or snacks.

Rigal (2012) Feeding Style and Feeding Strategy [123]

4

2: interviews with mothers

factor analysis

0.57-0.85

Construct Validity: partial least squares regression model identified several factors with regression coefficients >0.1 including: permissive style, contingency strategies, preference strategies, and coercion strategies.

Structural Validity: χ²(173) = 442.39, CFI = 0.88, NNFI = 0.86, RMSEA = 0.055.