FOR MORE, go to http://www.brookespublishing.com/piccolo

Technical Report

Gina A. Cook and Lori A. R ogg man

This technical report provides information on the development and psychometric properties of the observational measure Parenting Interactions with Children: Check- list of Observations Linked to Outcomes (PICCOLO). PICCOLO was developed at Utah State University in Logan, Utah, with funding from the Administration for Chil- dren, Youth, and Families of the U.S. Department of Health and Human Services (Grant #90YF0500) and from a Community University Research Initiative grant from Utah State University. After an overview of the instrument, the demographic characteristics of the samples used to develop PICCOLO are described. Then, the development of PICCOLO is discussed, including item selection, item analyses, and psychometrics. Next, descriptive statistics and information are provided about the measure’s reli- ability, including interrater reliability and scale internal consistency. Finally, valid- ity analyses are reported that include content validity ratings along with construct validity and predictive validity correlations.

OVERVIEW OF PICCOLO™

PICCOLO is an observational instrument to measure positive parenting. PICCOLO was developed to be easy to learn and practical to use by practitioners working with parents of young children. Psychometric data support PICCOLO as a measure that is reliable and valid. The PICCOLO domains are based on early child development theory and research suggesting critical dimensions of parent–child interactions that promote children’s development in social, language, and cognitive domains. These domains include parenting behaviors indicative of affection, responsiveness, en- couragement, and teaching. Indicator items in each of these domains were observed and evaluated on more than 4,500 video-recorded observations of parents interacting with their children ages 10–47 months. The PICCOLO domain indicators describe specific interactions between parents and children at home; scoring is not determined by the presence of specific toys or materials. This distinction between observed interactions and physical toys or materials is important because in some parenting programs, ma- terials and specific activities may be emphasized more than parent–child interac- tions. In PICCOLO, the focus is on what parents do with the materials they have and in the activities they engage in with their children.

Excerpted from PICCOLO User’s Guide by Lori A. Roggman, Gina A. Cook, Ph.D., Mark S. Innocenti, Ph.D., Vonda Jump Norman, Ph.D., Katie Christiansen, Ph.D., & Sheila Anderson, Ph.D. Brookes Publishing | www.brookespublishing.com | 1-800-638-3775 ' | All rights reserved

FOR MORE, go to http://www.brookespublishing.com/piccolo

46 Roggman et al.

SAMPLES FROM STUDIES PROVIDING OBSERVATION DATA

The information included in this technical report is drawn from extant observations and data from two research samples. Observations were not included if the family used a language other than English or Spanish or if the video recording was dam- aged, unclear, or missing. Each of these samples is described briefly next.

Early Head Start Research

Study summary: The Early Head Start (EHS) Research and Evaluation Project in- cluded 17 program sites from across the country and was funded by the Admin- istration for Children and Families in the U.S. Department of Health and Human Services to evaluate local EHS programs serving low-income families with chil- dren from before birth to age 3 years. This program included both center-based and home-based services.

Number of programs: 17

Number of participants: 3,001

Observation procedure: Children were observed with their primary caregiver at 14, 24, and 36 months of age. The observations consisted of a 10-minute, three-bag semistructured play interaction in which the parent was presented with three bags of play materials (the first bag with a book, the second bag with pretend toys, and the third bag with other toys) and asked to play with the child for 10 minutes using the bags in numerical order, dividing the time however they wanted. Of the 3,001 families (cases) in this study, 2,287 had one or more video- recorded observations of parent–child interactions (clips); of those, 1,986 fami- lies had at least one observation coded, with a total of 4,516 clips coded.

Bilingual Early Language and Literacy Supports

Study summary: The Bilingual Early Language and Literacy Supports (BELLS) proj- ect tested language and emergent literacy outcomes of bilingual children partici- pating in a program that serves low-income Hispanic children from birth through preschool in Utah. This program included both early English immersion and home language and literacy support.

Number of programs: 1

Number of participants: 112

Observation procedure: Children were observed with their primary caregiver at 14, 24, and 36 months of age. The observations consisted of a 15-minute two-bag play interaction (the first bag with books and the second bag with pretend toys); caregivers were asked to play with the child for 15 minutes using the bags in numerical order, dividing the time however they wanted. Of the 112 families in the sample, 62 had one or more video-recorded observations that were coded. (Attrition in this study was high due to family mobility.)

PICCOLO User’s Guide

PICCOLO™ SAMPLE DESCRIPTION Table A.1 shows the total number of families and observations from the samples described previously that were used to develop PICCOLO. These data are not intended to be nationally representative. Depending on the analytic question, either observations or families were used as the unit of analysis. The headings in the table refer to how that aspect of the sample was used in measurement development. Reliability N refers to the sample used to test interrater reliability across all observations scored by two or more observers. Scale N refers to the sample used to test scale reliability and construct validity across all observations scored by at least one observer. Validity N refers to the sample used to test predictive validity in cases with observations scored by at least two observers and with outcome data from children’s developmental assessments at a later time point. Descriptive N refers to the sample

developmental assessments at a later time point. Descriptive N refers to the sample used to describe the range, means, and standard deviations of PICCOLO domain and total scores at each age. The sample from the EHS study represents low-income families of European American, African American, and Latino American ethnicities who applied for the EHS program. The sample from the BELLS study provides additional observations from low-income families of primarily Latino American ethnicity. Demographic and descriptive data reported in this section come from questionnaires completed

and descriptive data reported in this section come from questionnaires completed by parents of these children.

graphic characteristics of the sample. Tables A.2 and A.3 show the proportion of family ethnicity at each level of maternal education and age.

MEASUREMENT DEVELOPMENT PROCESS

Table A.1. Sample of families and observations used to develop Parenting Interactions with Children: Checklist of Observations Linked to Outcomes (PICCOLO $ ^{\mathrm{TM}} $ )

TO TA L 1,823 2,048 2,369 aValidity N, number of families with at least one video observation coded by two observers and child out-

come data at a later age point. bD escriptive N, number of families with at least one video observation coded by at least one observer.

aValidity N, number of families with at least one video observation coded by two observers and child outcome data at a later age point.

48 Roggman et al.

Less than HS 31 48 HS or GED 39 29

Key: GED, obtained general equivalency diploma; HS, completed high school.

 20 years old

Item Development PICCOLO items were developed and identified initially by examining several sources. First, we examined definitions of constructs indicated in the research literature as linked to child outcomes and constructs central to major theories of child development. Second, we examined definitions of constructs we have found in our previous research to be linked to child outcomes in Head Start, EHS, and other lowincome populations. Third, we examined other existing observational measures of parent–child interaction, such as instruments widely used in the research literature. These instruments included Home Observation for Measurement of the Environment (HOME; Caldwell & Bradley, 1984); NCAST (Barnard & Kelly, 1990; Summer & Spietz, 1995); Maternal Behavior Q-set (Pederson & Moran, 1995); and assessment tools used by our Head Start program partners, such as the indicators described as interactive behaviors in the Hawaii Early Learning Profile (HELP; Parks, 1997), to identify behavioral indicators of positive parent–child interaction. Measures used for this purpose have established psychometric properties and are appropriate for low-income families from culturally and linguistically diverse groups. Finally, we reviewed with our Head Start program partners their program documentation, stated objectives, and intended outcomes for parent–child relationships. We used

be reworded to be concrete and parsimonious without losing their meaning. Items that could not be clarified in this way were eliminated. The remaining 112 items were selected for further work. The clarity and usefulness of behavioral indicators ultimately rests with those who use them to assess parent–child interaction. After initially reviewing various sources described previously for appropriate items and clarifying the wording of items to make them as clear and concrete as possible, we received feedback from

items to make them as clear and concrete as possible, we received feedback from untrained observers about both the clarity and importance of each item. Some be- Excerpted from PICCOLO User’s Guide by Lori A. Roggman,

FOR MORE, go to http://www.brookespublishing.com/piccolo

PICCOLO User’s Guide 49

haviors, even if clear and valued in the research literature, were neither clear nor important to nonresearchers. EHS staff and parent volunteers (n = 9) were asked to rate the large set of iden- tified and reworded items in two ways. How clearly is the item defined—Would you recognize the behavior if you saw it? How important is the item—How essential is it for children’s development? These ratings were used to identify items that needed to be clarified and to reduce the number of items to be tested in the initial version of the measure. To ensure that items are useful and appropriate for Spanish-speaking families, items were professionally translated into Spanish (with back translation) and then evaluated separately by a group of Spanish-speaking staff and parent vol- unteers (n = 4). After considering these ratings and reviews, 80 items were selected for test- ing through observations of archived video-recorded observations of parent–child interaction.

Item Selection

Teams of independent coders, recruited from undergraduate students in child de- velopment and psychology classes at Utah State University, tested the initial version of our checklist measure by coding observations from our video archive. Each stu- dent received 8–10 hours of training before observing video clips. Training included reading basic information about confidentiality protections, observational proce- dures, and parent–child interaction dimensions. To ensure that students understood the material they had read, each student was given a short quiz on the readings, and all students correctly answered at least 80% of the items before they were allowed to proceed. After training and certification in confidentiality procedures, students observed example video clips in small groups and discussed the observations with a member of the research team until reaching consensus. After passing a reliability test, students were assigned to code video clips. For assessing item reliability, two or more coders observed each video-recorded observation. For assessing predictive validity, at least two reliable coders observed each video-recorded observation. Multiple criteria were used to select the final PICCOLO items based on 1) vari- ability across individual cases, 2) reliability of raters, 3) scale reliability within do- mains, 4) construct validity with extant parenting measures, 5) predictive validity with extant child outcome data, 6) content validity from the importance ratings from program partners, and 7) qualitative feedback from raters and partners. Variability is shown in Table A.4.

RELIABILITY

Reliability refers to the degree to which the instrument is free from random error associated with the process of measuring the construct of interest. One step in mini- mizing random error in PICCOLO involves training materials that provide poten- tial observers with a clear and comprehensive understanding of the instrument’s purposes and procedures.

Training Observers

As part of the PICCOLO training, trainees read about the content and purpose of the measure (3 hours) and then watched and discussed five 10-minute video-

FOR MORE, go to http://www.brookespublishing.com/piccolo

50 Roggman et al. Table A.4. D escriptive statistics on Parenting I nteractions with Children: Checklist of O

14 months 7. 4 8 58.00 38.53 7. 6 6 24 months 12. 26 58.00 41.09 7. 7 7 36 months 6.09 58.00 40.54 7. 6 8

and domain in Tables A.5–A.8. The mean interrater correlation is shown for each domain and the total score in Table A.9.

aged together; data from child age 10 months not included. recorded segments previously observed by at least three expert observers who coded by consensus (3 hours). The consensus ratings established a standard by which to judge the accuracy of ratings made by trainees, and ratings that were not in agreement were then used to pinpoint additional training needs. At the end of training, potential users took a reliability test in which they watched and coded three to five additional 10-minute video-recorded observations to reach a level of

PICCOLO User’s Guide

Total = 7 items

Total = 7 items

FOR MORE, go to http://www.brookespublishing.com/piccolo

Roggman et al.

Encouragement.77 Teaching.76 PICCOLO Total.77

O utcomes. Each video clip was originally coded by a coder of the same ethnicity as that of the parent. Coders included European Americans, African Americans, and Latinos. Spanish-speaking families were observed by native Spanish-speaking coders. More than 500 observations were coded by two or more observers from a culture differ-

ment by culture is shown for each domain in Table A.10.

European A merican .81 A frican A merican.82 Latino.76

tions Linked to O utcomes. Excerpted from PICCOLO User’s Guide by Lori A. Roggman,

Scale Reliability

Scale Reliability PICCOLO was developed using a theoretical framework suggesting four major domains of parenting behaviors: Affection (warmth, care), Responsiveness (sensitivity), Encouragement (scaffolding, autonomy support), and Teaching (cognitive stimulation, language and literacy support). To test the degree to which data from actual parent–child interactions matched this theoretical framework, we conducted both internal consistency and confirmatory factor analyses on the data. Tables A.11– A.14 show the item descriptions and internal consistency statistics for items within each domain. Table A.15 presents the results of the confirmatory factor analyses. Within each domain, factor loadings are in the moderate to high range, consistent

with internal consistency within each domain. Given the factor loadings, the internal

7. Shows emotional warmth.63

8. R eplies to child’s words or sounds.41

FOR MORE, go to http://www.brookespublishing.com/piccolo

54 Roggman et al.

8. A sks child for information.51

8. A sks child for information.65 Excerpted from PICCOLO User’s Guide by Lori A. Roggman,

PICCOLO User’s Guide

R esponsiveness.61 Encouragement.73

Note: A ll correlations are significant at p < .001.

R esponsiveness.65 Encouragement.72

Note: A ll correlations are significant at p < .001.

R esponsiveness.65 Encouragement.71

in some domains. Tables A.23 and A.24 show t-test results from comparisons of

Note: A ll correlations are significant at p < .001. consistency within all four domains, and the strong theoretical basis of each of these domains, we recommend the use of all four domains. As shown in Tables A.16–A.18, PICCOLO domains are moderately to highly correlated with one an-

other, although less so over time and not at a level that would suggest that they measure the same construct.

Note: A ll correlations significant at p < .001. Excerpted from PICCOLO User’s Guide by Lori A. Roggman,

Roggman et al.

14 months 1.00 24 months.40

Note: A ll correlations are significant at p < .001.

14 months 1.00 24 months.44

p!<!.001.

Note: A ll correlations are significant at p < .001.

p!<!.001.

Note: A ll correlations are significant at p < .001.

domain scores between adjacent time points, with most changes occurring between 14 and 24 months.

VALIDITY

Encouragement 24 months 1.49 1,174.33 Teaching 14 months.93 1,176.35

PICCOLO User’s Guide

Encouragement 36 months 1.46 1,104.33 Teaching 24 months 1.09 1,104.36

validity, the extent to which practitioners in the field perceived the measure as including items that were important for parenting; construct validity, the relations between the PICCOLO measure and an established observational measure of par-

enting; and predictive validity, the association of the PICCOLO measure with positive child outcomes. The following aspects of validity were examined:

child outcomes. The following aspects of validity were examined: • Content validity from importance ratings by practitioners

were eliminated. Tables A.25–A.28 show the importance ratings for items in each domain.

Construct Validity Construct validity assesses the extent to which a measure is associated empirically

Project, an established reliable and valid observational parenting measure (Berlin, Excerpted from PICCOLO User’s Guide by Lori A. Roggman,

FOR MORE, go to http://www.brookespublishing.com/piccolo

58 Roggman et al.

7. Shows emotional warmth 3.00

7. R eplies to child’s words or sounds

FOR MORE, go to http://www.brookespublishing.com/piccolo

PICCOLO User’s Guide 59

Brady-Smith, & Brooks-Gunn, 2002; Fuligni & Brooks-Gunn, 2013) was used by a team of research scholars at Columbia University to code the same semistructured observations used for the development of PICCOLO. The dimensions of positive parenting in the established measure included Sensitivity, Cognitive Stimulation, and Positive Regard. Sensitivity was defined as the degree to which the parent was sensitive and child focused, provided praise and encouragement, and established a balance between giving support and allowing independent exploration. Cognitive Stimulation was defined as parent efforts to enhance perceptual, cognitive, and lan- guage development; to show awareness of the child’s developmental level; and to make efforts to bring the child above that level. Positive Regard was defined as the parent’s expressions of love, respect, and/or admiration for the child, including the quality and quantity of behaviors such as hugging, smiling, praising, and showing clear enjoyment of child. These three ratings were intercorrelated (r = .59–62; Berlin et al., 2002) and combined into one scale of Supportiveness for the national study (alpha = .82; Berlin et al., 2002) but were also used separately to examine the con- struct validity of specific PICCOLO domains. Reliability of the ratings was estab- lished at 85% agreement and maintained at 90%, allowing 1-point difference in scores (Administration for Children and Families, 2002). To examine the construct validity of PICCOLO, ratings of Sensitivity, Cogni- tive Stimulation, and Positive Regard were examined in relation to the PICCOLO domains of Responsiveness, Teaching, and Affection, respectively. The PICCOLO domain of Encouragement was also examined in relation to these dimensions of positive parenting available from the same observations but did not consistently overlap in definition with one of them. Encouragement and the PICCOLO total score were also examined in relation to the Supportiveness scale. Tables A.29–

A.31 show the associations at three age points between PICCOLO and the measure used to code the same observations in the EHS Research and Evaluation Project. Correlations are shown for European Americans, African Americans, and Latino Americans.

Predictive Validity

PICCOLO was designed to assess parenting behaviors that are directly associated with children’s positive outcomes, particularly the known indicators of school readiness. In this study, PICCOLO items and domains were significantly correlated with positive child outcomes within each ethnic group and across all three ethnic groups combined. All four domains were predictive of child development outcomes. When PICCOLO scores were high in any of the domains, children’s assessment scores were generally higher on measures of their cognitive and language development and often their social-emotional development as well. Table A.32 shows PICCOLO total and domain scores in relation to specific child outcomes for the combined sample and separate ethnic groups. To test the overall predictive validity of the PICCOLO scores, summary out- come variables were calculated using z-scores from outcome measures: a cognitive- language outcome variable (Peabody Picture Vocabulary Test–Third Edition [PPVT; Dunn & Dunn, 1997] and Bayley Scales of Infant Development: Mental Development Index [Bayley, 1993] at age 3, alpha = .72; PPVT and Woodcock-Johnson Psycho educational Test Battery–Revised: Letter Word and Applied Problems [Woodcock

Table A.29. Construct validity at 14 months for total sample, European Americans (EA), African Americans (AA), and Latino Americans (LA)

Excerpted 60 from PICCOLO Table A. 29.Construct validity at 14 months for total sample, European D omains A ffectio User’s Responsiveness Encouragement Teaching PICC O LO tota Guide Note: Key: by Lori A. Roggman,

Teaching.31.32.40.27.56.52.63.52.38.35.47.37.48.47 PICC O LO total.47.48.53.39.53.51.56.46.51.49.57.47.58.57

Table A.30. Construct validity at 24 months for total sample, European Americans (EA), African Americans (AA), and Latino Americans (LA)

Encouragement.47.37.57.28.39.34.47.21.42.30.48.37 Teaching.40.34.52.35.56.50.66.55.36.30.41.35 PICC O LO total.49.40.55.48.52.48.60.43.54.46.60.52

Excerpted from PICCOLO Table A. 30.Construct validity at 24 months for total sample, European D omains A ffectio User’s Responsiveness Encouragement Teaching PICC O LO tota Guide Note: Total sample ( Key: by Lori A. Roggman, 61

Table A.31. Construct validity at 36 months for total sample, European Americans (EA), African Americans (AA), and Latino Americans (LA)

Excerpted 62 from PICCOLO Table A. 31.Construct validity at 36 months for total sample, European D omains A ffectio User’s Responsiveness Encouragement Teaching PICC O LO tota Guide Note: Key: PICC O LO ™, Parenting by Lori A. Roggman,

Teaching.36.29.43.43.51.41.63.57.35.27.39.43.48.38 PICC O LO total.50.40.59.48.44.34.55.48.50.39.56.55.56.44

FOR MORE, go to http://www.brookespublishing.com/piccolo

(n!=!313-495)

(n!=!385!-!544)

.27^{x xtimes x x}M D124

.26^{\times x times times M}\ 124

..24^{x x x+}}\ \mathsf{M D}\ vert,6

.20^{\ \ \times times\times\ \ \times}\ {\sf M D}\vert36

.17^{\ \ \ \times\times\times}\ \ \ \ !,2^{\circ}

.18^{x\times x\times\ }\mathsf{P P V}\mathsf{T}36

.18^{\ \ \times\times\times\times}M D\ 24

.13^{\ast}\mathsf{C D}|,24^{}

.99^{\ \ \times\times\times\times}M D/36

.12^{****}\text B R S24

.15^{*},R R S,24

.15^{\ast},\mathsf{B R S}36

.12^{\star\ast},\mathsf{B R S24}

.22^{x\times x\times}\mathsf{P P V}\lceil\ 6^{\circ}

.21^{x\times x\times\ }\mathsf{P P V}\mathsf{[T}\mathsf{36}

.26^{\ \times*}\sf{P P V T T P K}^{}

.16^{x x}}\ \times\times\ \cup{A P P K}

.23^{\times\times\times\times}\mathsf{P P V T P K}

.17^{\ *},{mathsf P P V},{\mathsf P K}

.13^{\star}\mathsf{W J A P,K K^{a}}

.20^{x+x+x}W N P P K

.99^{x+x+y}W cup A P P K^{a}

.88^x+x+times M vert136

.23^{xtimes x x+}\mathsf{M D}\vert36

.17^{x\times x\times\ }M\ \ !1,36

-.16^{\times\times\times\times\times}\ C B C P K

25^{x+x x+}\mathsf{P P V}\uparrow36^{\circ}

.21^{\ \ \ \ \

.14^{\star\star\star},\mathsf{P P V T},36

330^{x\times x\times}\mathsf{P P V}\mathsf{P K}^{a}

.16^{\ \ \times\times\times\times}\mathsf{P P V T P K}

.22^{\times\times\times\times}\mathsf{P P V T}P\mathsf

.21^{\ \ \times times\timestimestimes\ \cup}P P P K

.27^{*****\mathsf}\mathsf{P P V T P K}^{\ }

.21^{\ \ \times\times\times\times}W\cup{\\ \ }}\ \end{array*

.16^{\ \ \times\times\times\times},\mathsf{P P V T P K}

.6^{times times\times\times}W\cup A P P K^{circ}

.16^{\times*times}\mathsf{W J P P P K}

.22^{\times x times x times}\ M D\vert24

.9^{x+x+x}M D/24

..25^{\times x\times x\times}\ {sf M M}\ vert,6

.18^{x\ \ \ \times\ \times\ \times}\ \mathsf{M D}\ vert6

.20^{\ \ \times\times\ }M\ 1124

.5^{***},\ \ {\mathrm M{}}\ !\ !\

.17^{\times x times\times x}\ \ .

.44^{x\ x\ x\ *}\ \text{cled D}\ 24

.66^{\ \ \ast\ast\ast\ast}\ \ {scriptstyle D}\vert24^{\circ}

.21^{x\ x\ \ \ \times\ }\mathsf{P P V T}P K^{a}

.14^{\ \ \\ \ }}mathsf{W J L M}\ {K^{\ \ }}

.44^{\ \ \times\times\ }R S24

.16^{\ \ \ \times\times\timestimestimes\ }\sf{P P V T}}\uparrow36^{\circ}

.20^{x xtimes x}\mathsf{P P V T36}

.19^{*****}\mathsf{P P V T P K}

.15^{***},\mathsf{P P V T/36}

.20^{\ \ \times\times\ }\mathsf{P P V T P K}

\ {\begin{array}{l}{15^{****}W!{\cal J}\ {\cal M}!{\cal P}!{\cal K}}\ \end{array}}

.18^{\ \ \times\times\times\times\times}N\cup\cup{\cal L}W P\

.16^{**\times*}\mathsf{W J A P P K}^{mathsf a

.15^{****}\mathsf{W J A P P K}

.27^{\times x\times x}\mathrm{M D}\vert36

.22^{\ ****,}\mathsf{P P V T},6^{\circ}

.33^{\times x times x times}

.19^{\ \times*}\mathsf{W}\mathsf{J}P P\mathsf{K}

.24^{\ \ \times\times\times\ast},\mathsf{P P V T T K^^{\ },}}

.18^{\ \ \times\times\times\times}\ \ {sf M M}\vert\,,{36}

-.16^{\ \ \\ \times\times\times}\mathsf{C B C}36

.21^{*****,}\mathsf{P P V T P K}

.15^{\timestimes}\mathsf{W J L W}\mathsf{P K}^{\mathsf{a}}

-.13^{ast}\mathsf{C B C P K}

.27^{\ \ times\times\times\times}\sf P V V T\3,3

.19^{*****}\mathsf{M D I}36

-.13^{\times*\times},\mathsf{C B C},36

.14^{\ \ \times\ }\mathsf{W J A P P K}

.20^{x x+times x}\ {P}}{}\ {

.44^{x\times x\times x}\ {sf N J L W}P P^{a}

.23^{x\times x\times}\mathsf{P P V}\mathsf{P K}

.19^{**},\mathsf{B R S},24

.20^{\ \ \ \times\times\ast},\mathsf{P P V T},\mathsf{P K}

Encouragement

.13^{\ast},\mathsf{W J L W},\mathsf{P K}

.23^{****}\mathsf{P P V T}P K^{\mathsf{a}}

.26^{\ \ \ \times\times\times}N=322W,L

.8^{***\times*}\mathsf{W J A P P K}

25^{x\pm x\pm N}=424\ !!P P P!

.21^{\times}\mathsf{P P V T T}\mathsf{P K}

.17^{**}\mathsf{{*}}\mathsf{{W L W P}}}\mathsf{{{K}}}

FOR MORE, go to http://www.brookespublishing.com/piccolo

64

(n!=!385!-!544)

.22^{\times x times times}\ \ !!{D}24

23^{x+x+x}M D\mid24

.18^{x\times x\ times}}\ \mathsf{M D}\ vert,,6

.18^{**},Rmathsf S4

.27^{\times x times x times}/D124

.13^{****}\big{\big{Dbig}big}\big|,24^{circ

.25^{\ \ \times times\timestimes}/D\ \ 36

20^{x\times x\times x}\ \mathsf{P P V}\mathsf{I7}}36^{\circ}

.28^{x\times x\times}B R56

26^{x+x x+}\mathsf P P V\ !6

.17^{\ \ \times times\timestimes\ \ \ {Psf V T}K}

.20^{x\times x\times}\mathsf{P P V T}\mathsf{P K}^{a}

.18^{**},\mathsf{P P V T},36^{\circ}

.17^{x times x+}\left{\sf C D}\right|vert\ 24

.66^{\ \ \times\times\times\ }B R S24

22^{x+times x+}\mathsf{P P V}\ !6

15^{\ \ \times\times\times\times\times\cup\cup/N P K^{\circ}}

.13^{\ \times\times\times\times}\ \cup!!!!\sf{A A P P K}^{}

.12^{\ast}\mathsf{W J L W},mathsf P P K

.29^{\ \ \ \times\times\ }\mathsf{P P V T P K}

.27^{\times x+x}M D/36

.44^x x+x M M D\ 36

.21^{\ \ \times\times\times\ }\ \mathsf{W J L W P K}

.18^{x x+x x}/N\ P P P

.0^{x\times x\times x}\mathsf{P P V}\mathsf36^{a circ

.18^{\ \ \ast},\ \ \mathsf{M D}\vert,36

-.18^{\ \ \times\times\timestimes\ },{\sf C B C P K}

.25^{x x x}}

.24^{\ \ \times\times\times}\sf P V V T\,K K^{\ }

.16^{*},\mathsf{B R S},36

24^{x+x x+}P P V736^{a}

.18^{\ \times\times\times\times}W\ L W P K^{a}

.20^{\ \times*}\mathsf{P P V T K}

.33^{*****,}\mathsf{P P V T},\mathsf{P K}

.16^{\ \ \times\times\timestimestimes\ }\mathsf{W}\mathsf{J}P P\mathsf{K}^{}

.16^{***}W\ W P K

.44^{*}\backslash{\cal W}\backslash{\cal L}P K^{\ }!

.12^{**}\mathsf{W J A P P K}

.22^{x\times x\times x}\ N\cup N P\times

.15^{\star\star}\mathsf{W J A P P K^{\circ}}

.22^{\ \times\times\times\times},\mathsf{P P V T P K}

.24^{\ \ast\ast\ }\mathsf{P P V T T P K}^{\mathsf}a^

.30^{***},\mathsf{P P V T T P K}

.22^{x x+x x}\ N\cup

.31^{\times}\mathsf{P P V T}P K

.21^{\ \star\times\ }\mathsf{W J L W P K}

15^{\ \times\times\times\times}W\cup A P P K^{\circ}

.17^{\ *\ast}\mathsf{W J L W P K}{}^{\ 2

.21^{x x+x x}/N\ A P P K

.17^{\ \ \times\times\times\times}\mathsf{W J A P P K}

.21^{x x+x x}M M124

.33^x x x+\ D\ \ 24

.18^{**},\mathsf{B R S}{24}

.23^{\ \ \times times\timestimes\ \ /}\ !,

.27^{\times x+x M D}\ 24

.22^{\times x times\times}\ \mathsf{M D}\ vert,66

.25^\times\times\times\times times\ \ {sf M M}}\ 1,6

.13^{\ \ast\ast}\left langle!,!24^{\circ}

.17^{\ \ \times\times\times\times}\ \ .\ \ .

.28^{\ \times\times*}\mathsf{B R S36}

.13^{\ \ \ast\ast\ast}\ \ !,!,!,!,24

24^{\ x x\times x\times}\mathsf{P P V}\mathsf{T}\mathsf{36^{a}}

.18^{\ **}P P V/36^{\circ}

.26^{\ \ \ \times\ \*\times},\mathsf{P P V T},\ 6

.16^{x+x+x}

.22^{\times times\times\ast}\sf P V V T\33

.21^{\ \ \ \times\ \\ \times\ }\mathsf{P P V T P K P}^{}

.9^{\times*\times*},{\sf P P V T},{\sf P K}

.18^{*},\mathsf{P P V T K}

.21^{\ \ \ \times\ \*\times}\mathsf{W J L W},\mathsf{P K}

.55^{\ \ \times\times\times\ }\ \mathsf{W J L W K P P}^{\ }

.12^{\star}\mathsf{W J L W},mathsf P P K

.27^{xtimes x x+}D/36

..25^{\times x\times x\times}}M\ \ \vert,{36}

.88^{\ *\ast},\ \ {\sf M D}\vert,6

.25^{\times x times\times}\ {sf M M}\ vert,6

aPartial correlation controlling for test language (English or Spanish). & Johnson, 1989] at prekindergarten, alpha = .74) and a cognitive-language-social outcome variable (adding Bayley Scales of Infant Development: Behavior Rating Scales [Bayley, 1993] and reverse-scored Child Behavior Checklist [CBC; Achenbach & Rescorla, 2000] Aggression at age 3, alpha = .64; and CBC Aggression at

-.18^{\times\times}\mathsf{C B C P}

.24^{x\times x\ \times}\sf P V V T\336^{a}

.16^{*},\mathsf{B R S},36

.23^{****}\mathsf{P P V T}P K^{a}

-.12^{\star\star},{\sf C B C},36

.26^{x\times x\times}\sf{P P V T66

\begin{aligned}{}&{{}10^{\pm\times\ast\mathsf{P P V}}7766}\ {}&{{}20^{\pm\times\ast\ast\mathsf{P P V T}}6}\ \end{aligned}

\ _{{cdots}}6^{***\mathrm{W}}\mathsf{J J L W P K}

.23^{\\ \times\times\times\times},\mathsf{P P V T P},\mathsf{P K}

.16^{\ \ \times\times\times\ }}\mathsf{W}\mathsf{J}\mathsf{P}\mathsf{P}\mathsf{K}^{}

.21^{**\times*}\mathsf{W J L W},{\mathsf{P K}}

.44^{*}\mathsf{W L L W P P}^{}!

.55^times M\cup A P P K^{\ }nonumber

.22^{x\times x\times x}\mathsf{W J A P P K}

.25^{****}\mathsf{P P V T T P K}^{\mathsf{a}}

.21^{****}\mathsf{W J A P P K}

.31^{*****\verb}P\mathsf{P P V T P K}

.24**** WJLW PKa.21**** WJLW PK.17** WJLW PKa.27**** WJLW PK .16**** WJA P PKa.17**** WJA P PK.21**** WJA P PK Note: Correlations are included in the table if p < .05 and r > .11. Key: BR S, Bayley Scales of I nfant D evelopment: Behavior R ating Scales: Emotion R egulation (24 months, 36 months; Bayley, 1993); CBC, Child Behavior Checklist A ggression score (24 months, 36 months; A chenbach & R escorla, 2000); CDI munication D evelopment I ndex: Vocabulary Production (24 months; Fenson et al., 1994); MDI, Bayley Scales of I D evelopment: M ental D evelopment I ndex: Cognitive D evelopment (24 months, 36 months; Bayley, 1993); PPVT, Peabody Picture Vocabulary Test–Third Edition: R eceptive Vocabulary (36 months, prekindergarten; D unn & D unn, 1997); WJA Woodcock-Johnson Psychoeducational Test Battery–R evised: A pplied Problems: Problem Solving (prekindergarten;

FOR MORE, go to http://www.brookespublishing.com/piccolo

PICCOLO User’s Guide

.25^{**}

.25^{**}

PI CCO LO 2 years.27.27.24 PI CCO LO 3 years.21**.24**.19

**p < .01; partial correlations, controlling for testing language (Spanish or English). Statistically significant correlations between these constructed outcome variables and PICCOLO at each age, as well as PICCOLO averaged for the three ages,

ables and PICCOLO at each age, as well as PICCOLO averaged for the three ages, demonstrated that the PICCOLO measure predicts children’s developmental outcomes (see Table A.33). The psychometric properties of the PICCOLO measure have been tested and show considerable evidence of multiple aspects of reliability and validity. The items that make up PICCOLO can be reliably observed by non-experts with only a few hours of training. The four domains of PICCOLO—Affection, Responsiveness, Encouragement, and Teaching, represent reliable scales. The items show content validity, and the domains and total PICCOLO scores show construct validity. Finally, the domains and the total PICCOLO scores predict positive outcomes for children, particularly the cognitive, language, and social skills that underlie school

Excerpted from PICCOLO User’s Guide by Lori A. Roggman,