Picture My Participation (PMP) intended to measure participation, defined as attendance and involvement in everyday situations, of children with disabilities, particularly in low- and middle-income settings.
To explore structural validity of PMP by identifying possible subcomponents in the attendance scale and examining internal consistency of the total score and each subcomponent.
A picture-supported interview was conducted with 182 children, 7–18 years, with and without intellectual disability (ID). Frequency of attendance in 20 activities was rated on a four-point Likert scale (never, seldom, sometimes and always).
An exploratory principal component analysis extracted four subcomponents: (1) organised activities, (2) social activities and taking care of others, (3) family life activities and 4) personal care and development activities. Internal consistency for the total scale (alpha = 0.85) and the first two subcomponents (alpha = 0.72 and 0.75) was acceptable. The two last subcomponents alpha values were 0.57 and 0.49.
The four possible subcomponents of PMP can be used to provide information about possible domains in which participation and participation restrictions exist. This study provided further psychometric evidence about PMP as a measure of participation. The stability and the utility of these subcomponents needed further exploration.
The participation construct is considered as an essential reflector of an individual actual function in real life and should therefore be researched in clinical practices and specifically in relation to disability and health (United Nations
When endorsing an integrative and multidimensional understanding of functioning and health for individuals with disability, participation can be considered as a reflector of the interaction between body impairments and societal barriers (Arvidsson et al.
Picture My Participation (PMP) is a self-report instrument that was specifically designed to capture the two aspects of participation, namely, attendance and perceived involvement in children and youth with mild ID in 20 different activities related to home, social life and community (Arvidsson et al.
This cross-sectional, instrument validation study was designed according to the COnsensus-based Standards for the selection of health Measurement INstruments (COSMIN) principles (Mokkink et al.
To obtain data on the utility of the PMP measure for countries with different cultures and income levels, data were collected in South Africa, Taiwan, Mainland China and Sweden. In South Africa, the study was conducted in a city of approximately 200 000 inhabitants, in Taiwan in one city with approximately 2.6 million inhabitants, in Mainland China in two cities with 7.6 and 15.6 million inhabitants, respectively, and in Sweden in two cities with approximately 100 000 inhabitants each.
An instrument with universal utility is valid and reliable under different circumstances. The purpose of the sampling strategy was to ensure variation in the samples in terms of age, gender, country/context, socio-economic circumstances and level of disability. In addition to targeting samples from four different countries, we sought children with mild ID from all four countries. Children with typical development (TD) were for reasons of convenience recruited only in South Africa. Children with TD were recruited to obtain variation of participation also within low- and middle-income setting. Consequently, five subgroups of children were recruited: (1) children with ID in South Africa (
Descriptive data regarding gender, age and Picture My Participation total scores for the five subsamples and for all participants together.
Variables | Five subsamples |
All participants together ( |
|||||
---|---|---|---|---|---|---|---|
Children from South Africa with ID ( |
Children from South Africa with TD ( |
Children from Mainland China with ID ( |
Children from Taiwan with ID ( |
Children from Sweden with ID ( |
|||
Gender, |
Girls | 36 (46%) | 22 (67%) | 8 (40.0 %) | 11 (36.7 %) | 6 (30.0 %) | 83 (45.6 %) |
Boys | 38 (48%) | 11(33%) | 12 (60.0%) | 19 (63.3 %) | 14 (70.0 %) | 94 (51.6 %) | |
Missing | 5 (6%) | 0 | 0 | 0 | 0 | 5 (2.7 %) | |
Age (years) | Min – Max | 9 – 16 | 9 – 14 | 7 – 18 | 8 – 12 | 7 – 18 | 7 – 18 |
Mean (s.d.) | 12.7 (1.7) | 11.2 (1.6) | 12.3 (3.1) | 10.5 (1.3) | 11.7 (3.1) | 11.9 (2.2) | |
Missing (single items) | 5 | 0 | 0 | 0 | 0 | 5 (2.7 %) | |
PMP total score (19 items) (Likert scale: 1–4 Score range: 19–76) | Min – Max | 28 – 74 | 48 – 69 | 30 – 66 | 49 – 63 | 28 – 69 | 28 – 47 |
Mean (s.d.) | 53.2 (11.4) | 60.8 (5.7) | 47.5 (7.5) | 49.4 (5.7) | 51.3 (9.5) | 52.4 (8.0) | |
Missing (single items) | 2 | 1 | 0 | 0 | 7 | 10 |
PMP, Picture My Participation; ID, intellectual disability; TD, typical development.
Children were eligible for inclusion if they had been diagnosed with ID and attended a school for children with ID, as confirmed by their caregivers. Children with either ID or TD also needed to meet the following criteria to be included: (1) aged between 7 and 18 years, (2) able to speak and understand English (in South Africa), Swedish (in Sweden) or Mandarin (in Mainland China and Taiwan), and (3) assented to participate in the study. For all children, the legal caregiver had to give consent for their child to participate.
All the data were collected by clinical researchers who conducted structured interviews or by specially trained postgraduate students with knowledge about the target group and the PMP. Data related to participant characteristics, including date of birth and gender, were collected using a parent-report survey.
The participation instrument PMP (Arvidsson & Granlund
frequency of attendance for each item, rated on a four-point Likert scale (never, seldom, sometimes and always)
selection of the three most important activities according to the child
perceived involvement (by the child) in these three activities, rated on a three-point Likert scale (not involved, somewhat involved and very involved). In this section, the children were also asked if there was any other activity that they would select as important, besides the 20 activities that were asked about in the PMP
evaluation of perceived barriers to and facilitators of participation in relation to the activities that were the most important to the children.
Descriptive statistics of the principal component analysis based on all Picture My Participation items.
Variables | Frequencies of ratings |
Mean score | s.d. | ||||
---|---|---|---|---|---|---|---|
Always (4) | Sometimes (3) | Seldom (2) | Never (1) | Missing N | |||
Personal care | 136 | 32 | 4 | 4 | 0 | 3.7 | 0.63 |
Family mealtime | 106 | 54 | 8 | 8 | 0 | 3.5 | 0.79 |
My own health | 47 | 44 | 48 | 37 | 0 | 2.6 | 1.10 |
Gathering supplies | 35 | 57 | 31 | 52 | 1 | 2.4 | 1.11 |
Meal preparation | 29 | 48 | 38 | 59 | 2 | 2.3 | 1.10 |
Cleaning at home | 45 | 73 | 33 | 24 | 1 | 2.8 | 0.98 |
Caring for family | 56 | 52 | 27 | 39 | 2 | 2.7 | 1.13 |
Caring for animals/pets | 46 | 22 | 30 | 78 | 0 | 2.2 | 1.26 |
Family time | 90 | 49 | 27 | 10 | 0 | 3.2 | 0.92 |
Celebrations | 57 | 63 | 39 | 17 | 0 | 2.9 | 0.96 |
Playing with others | 50 | 63 | 30 | 33 | 0 | 2.7 | 1.07 |
Organised leisure | 65 | 46 | 34 | 31 | 0 | 2.8 | 1.12 |
Quiet leisure | 78 | 44 | 33 | 21 | 0 | 3.0 | 1.06 |
Spiritual activities | 57 | 39 | 40 | 40 | 0 | 2.6 | 1.16 |
Shopping | 46 | 67 | 35 | 27 | 1 | 2.8 | 1.01 |
Social activities | 13 | 44 | 38 | 78 | 3 | 2.0 | 1.00 |
Health centre | 28 | 72 | 57 | 19 | 0 | 2.6 | 0.88 |
School | 131 | 33 | 4 | 8 | 0 | 3.6 | 0.74 |
Overnights visits and trips | 52 | 68 | 39 | 17 | 0 | 2.9 | 0.95 |
s.d., standard deviation.
Administration took 20 to 30 min for each child. For the purposes of this study, only data from section 1 (frequency of attendance) were used.
The PMP was completed as part of structured interviews in which graphic symbols from the aided symbol set of Picture Communication Symbols (PCS™) were used (Fuller & Lloyd
Participant characteristics were summarised descriptively. The four-point Likert scale for measuring attendance was prepared with the following values: 1 = never, 2 = seldom; 3 = sometimes, 4 = always; total scores were calculated by averaging responses to each item. Total scores were summarised descriptively for each subsample and for the total sample. All statistical analyses were performed using SPSS 24.0.
An exploratory principal component analysis (PCA) was used as the extraction method to explore the dimensionality of the scale and investigate possible subcomponents of the PMP. In this PCA, 19 of the 20 items were used. The item ‘paid and unpaid employment’ was excluded based on the experience of the data collectors. No child was attending any activity that could be considered as employment and most of the children were confused by the question. The rotation method used was Varimax with Kaiser Normalisation and the result was that the rotation converged in nine iterations, eigenvalues >1.
Children’s attendance in different activities tends to vary between type of activities with certain types being more commonly attended universally (e.g. within family activities), whilst others may vary depending on economical circumstances (e.g. organised leisure activities outside home) (Arvidsson et al.
Cronbach’s alpha coefficients were used to calculate the internal consistency of the total scale of the PMP and for any identified subcomponents. Alpha values greater than 0.70 are considered to demonstrate adequate internal consistency (Terwee et al.
Ethical approval for the study was obtained from the Ethics Committees and Boards in each of the four participating countries (University of Pretoria , South Africa, reference number: GW20180301HS; Tianjin Medical University , People Republic of China, reference number: TUMEC20140201; Regionala etikprovningsnamnden, Sweden, reference number: 2017/234-32; Chang Gung Medical Foundation, IRB number: 201600861B0) and from the relevant local Departments of Education and school principals. Informed consent was obtained from every child’s primary caregiver and consent was also sought from every participating child in each of the countries involved in the study.
Participant characteristics are presented in
Principal component analysis extractions with total variance explained, with initial eigenvalues and after rotation, are presented in
Component matrix demonstrating four components that are evident in the data set. Scree plot.
Principal component analysis extraction with total variance explained, with initial eigenvalues and after rotation.
PMP item | Initial Eigenvalues |
Rotation sums of squared loadings |
||||
---|---|---|---|---|---|---|
Total | % of variance | Cumulative % | Total | % of variance | Cumulative % | |
1 | 5.1 | 26.7 | 26.7 | 2.9 | 15.1 | 15.1 |
2 | 1.5 | 7.9 | 34.6 | 2.8 | 14.7 | 29.7 |
3 | 1.4 | 7.3 | 41.8 | 1.9 | 9.9 | 39.6 |
4 | 1.3 | 6.7 | 48.5 | 1.7 | 8.9 | 48.5 |
5 | 0.9 | 5.0 | 53.5 | - | - | - |
6 | 0.9 | 4.8 | 58.2 | - | - | - |
7 | 0.9 | 4.7 | 62.9 | - | - | - |
8 | 0.9 | 4.6 | 67.5 | - | - | - |
9 | 0.8 | 4.1 | 71.5 | - | - | - |
10 | 0.8 | 4.0 | 75.5 | - | - | - |
11 | 0.7 | 3.5 | 79.0 | - | - | - |
12 | 0.6 | 3.4 | 82.4 | - | - | - |
13 | 0.6 | 3.0 | 85.5 | - | - | - |
14 | 0.6 | 3.0 | 88.5 | - | - | - |
15 | 0.5 | 2.6 | 91.1 | - | - | - |
16 | 0.5 | 2.6 | 93.7 | - | - | - |
17 | 0.4 | 2.2 | 95.9 | - | - | - |
18 | 0.4 | 2.1 | 98.0 | - | - | - |
19 | 0.4 | 2.0 | 100.0 | - | - | - |
PMP, Picture My Participation.
Principal component analysis – Rotated component matrix.
Variables | Component |
|||
---|---|---|---|---|
1 | 2 | 3 | 4 | |
Overnight visits and trips | 0.702 |
- | - | - |
Organised leisure | 0.654 |
- | 0.203 | - |
Cleaning at home | 0.635 |
0.120 | 0.143 | 0.205 |
Health centre | 0.594 |
0.348 | -0.126 | - |
Gathering supplies | 0.585 |
0.120 | 0.185 | 0.106 |
Shopping | 0.456 |
0.312 | 0.173 | 0.154 |
Playing with others | - | 0.678 |
0.113 | 0.131 |
Caring for family | 0.129 | 0.635 |
- | 0.300 |
Spiritual activities | 0.138 | 0.621 |
0.241 | - |
Celebrations | 0.210 | 0.571 |
- | 0.257 |
Caring for animals/pets | 0.413 | 0.523 |
-0.267 | - |
Social activities | 0.181 | 0.490 |
0.344 | - |
Meal preparation | 0.144 | 0.490 |
0.436 | - |
Family time | - | 0.355 | 0.685 |
- |
Family mealtime | 0.175 | - | 0.649 |
0.396 |
Quiet leisure | 0.402 | - | 0.599 |
|
School | - | - | 0.152 | 0.741 |
Personal care | 0.103 | 0.108 | - | 0.721 |
My own health | 0.355 | 0.236 | - | 0.414 |
, statistically significant.
This subcomponent includes the following six activity items: trips and visits, organised leisure, cleaning at home, health centre (visits to), gathering supplies and shopping. It involves events or pursuits that a group of people are doing together in a structured way. This implies that there is a collective structure to the activity rather than it being performed as an individual activity.
This subcomponent includes the following seven activity items: playing with others, caring for family, spiritual activities, celebrations, caring for animals or pets, social activities and meal preparation. It involves events or pursuits that bring members of the community together.
This subcomponent includes the following three activity items: family time, family mealtime and quiet leisure. It involves events or pursuits that bring members of the family together.
This subcomponent includes the following three activity items: school, personal care and my own health. It refers to both basic self-care tasks of bathing, dressing, personal hygiene and grooming, as well as more complex tasks related to health and education.
The internal consistency calculated by Cronbach’s alpha for the total scale was 0.85. Cronbach’s alpha for the subcomponent
The main aim of this study was to explore the structural validity of the PMP instrument. The results of the PCA identified four subcomponents, namely,
In addition to the relatively high structural stability (high component loadings) that the items had in these subcomponents, they also made sense clinimetrically. These two subcomponents may consist of items that from a clinical perspective are interrelated (non-routine activities taking place in the home and personal routines), but they may not have strong relationships at an item level.
When evaluating scale properties using both statistical and theoretical perspectives, often, there are choices to be made that can influence which items will be clustered together into a scale. In this study, a four-component solution in the PCA was supported by both theoretical and statistical perspectives. Statistically, a two-component solution is probably as good as a four-component solution (see scree plot,
Firstly, the subcomponent
Secondly, the subcomponent
Thirdly, the subcomponent
The last subcomponent,
The PMP was used to gather the child respondents’ own views about their participation in everyday activities. In developing the instrument, special focus was placed on making sure that children with cognitive problems, which might affect their understanding of items and scales in a questionnaire, could participate in the study. This was established by the three trial items in first step of the PMP procedure. This procedure facilitated the establishment of the children’s understanding of the concepts and of understanding the instructions. Another focus was to make sure that the items asked were relevant in low-resource settings and for varying cultural groups. For this reason, heterogeneous groups of children were sought for the validation: those with ID, TD, from different countries and from both high-income and low- or middle-income settings, across a fairly broad age group (7 to 18 years). However, another less explicit but shared characteristic was that all the participants, in different ways, experienced participation and participation restrictions in their everyday lives. The need and right to experience participation can be considered universal amongst all children and according to the WHO and United Nations Childrens Fund (UNICEF), all children also have the right to express their perceptions of such participation (Carroll-Lind et al.
In designing the study, consideration was given to reducing the risk of bias, including sampling adequacy (sample size > 100 and seven times the number of items), appropriateness of the analysis methods and clarity of description of procedures (Mokkink et al.
This study contributes evidence of validity in children with mild ID: future directions include testing validity in other disability groups. It is challenging to assess participation in children with disabilities (Coster & Khetani
In this study, the structural validity of the PMP was explored by identifying possible subcomponents in the attendance scale and examining internal consistency of the total score and of each subcomponent. An exploratory PCA extracted four subcomponents: organised activities, social activities and taking care of others, family life activities and personal care and development activities. Internal consistency for the total scale and the first two subcomponents were acceptable. The four subcomponents of PMP can be used to provide information about possible domains in which participation and participation restrictions occur.
The authors declare that they have no financial or personal relationships that may have inappropriately influenced them in writing this article.
P.A., S.D., M.G., C.I., L.J.S., L.J.K., A-W.H and K.H. all contributed equally to this work.
This project was jointly funded by the South African National Research Foundation (NRF) (101566) and the Swedish Foundation for International Cooperation in Research and Higher Education (STINT) (SA2015-6253).
The data that support the findings of this study are available from the corresponding author, S.D., upon reasonable request.
The views and opinions expressed in this article are those of the author and do not necessarily reflect the official policy of the affiliated agency of the authors or the funders.