Environmental Restriction Questionnaire: cross-cultural adaptation, reliability and validity of Turkish version in young children with developmental disabilities

Abstract

Background

Children with developmental disabilities often face environmental barriers that limit their participation in daily activities. Environmental Restriction Questionnaire (ERQ) requires cross-cultural adaptation and validation to be used in Turkish populations. The aim of this study was to evaluate cross-cultural adaptation, reliability, and validity of the Turkish version of the Environmental Restriction Questionnaire (ERQ) in young children with developmental disabilities.

Methods

Participants were a total of 220 children, aged between 4 and 6 years, who had developmental disabilities (mean age: 4.95 ± 0.9) and their parents. The Environmental Restriction Questionnaire (ERQ) was translated from English into Turkish and underwent cross-cultural adaptation. Construct validity was evaluated using confirmatory factor analysis, and concurrent validity was determined using the correlation between ERQ and CASP with Spearman’s correlation analysis. For reliability of ERQ, Cronbach’s alpha (α) and Intraclass Correlation Coefficient (ICC) were used.

Results

For reliability of ERQ, high Cronbach’s α and intraclass correlation coefficient (ICC) values were found for the ERQ total and subscale scores (ICC = 0.80–0.90) (Cronbach’s α = 0.84–0.93). Construct validity showed poor model fit (Root Mean Square Error of Approximation (RMSEA) = 0.112; Goodness of Fit Index (GFI) = 0.648) suggesting that the proposed factor structure was not adequately supported. Concurrent validity analyses revealed weak to moderate negative correlations between ERQ and CASP scores (r = − 0.351 to − 0.135; p < 0.05).

Conclusion

The ERQ appears to be a reliable, cross-culturally adapted instrument; however, its structural validity was not adequately supported, and its applicability should be interpreted with caution. While the Turkish version of the ERQ demonstrates good reliability, construct validity analysis indicated poor model fit, suggesting that the factor structure requires further investigation.

Introduction

Early childhood is a critical period characterized by rapid growth, during which the developing brain is highly sensitive to environmental stimuli and nurturing interactions. This phase plays a crucial role in shaping children’s developmental outcomes. Early development depends on interactions with parents and family members and is reinforced by a supportive environment that facilitates these interactions [1]. Compared to their typically developing peers, children with disabilities, tend to engage in activities that are more limited in scope and lower in intensity, and are often individual, home-based, and sedentary activities. As a result, they may be deprived of the various benefits associated with participation [2].

The prevalence of childhood disability in Turkey is above the global average [3]. The systematic assessment of environmental factors and dimensions of participation has become a critical requirement for enhancing children’s developmental outcomes, quality of life and social participation [4–7]. Although, in the literature, various instruments have been proposed to measure participation and environmental restrictions, it has been emphasized that these need to be adapted so that they are appropriate to the cultural context [8]. In Turkey, in particular, children with developmental disabilities are affected by the limited availability of valid and reliable instruments for assessing environmental restrictions and participation in daily life activities, which underscores the importance of cross-cultural adaptation studies in this field. To the best of our knowledge, a very limited number of tools is available in the literature for assessing environmental factors in children. One such tool, The European Child Environment Questionnaire (ECEQ) was developed to assess environmental factors in children with cerebral palsy, but not developed specifically for early childhood. Among the strengths, instruments such as the ECEQ offer a detailed evaluation of environmental factors. Construct validity of the ECEQ showed mixed model fit, with acceptable GFI values (≥ 0.90) but elevated RMSEA values (> 0.08), indicating limitations in model fit. However, concurrent validity (r = − 0.533 to − 0.293; p < 0.05) and high internal consistency (Cronbach’s α = 0.82–0.96) were demonstrated in Turkish samples [9, 10]. The Home Observation for Measurement of the Environment (HOME) is another tool that assesses environmental elements, but only within the child’s home setting rather than the wider environment. The internal consistency of the total scores shows about 0.8, while the consistency of a single subscale ranges from 0.3 to 0.8 [11, 12]. Similarly, the Participation and Environment Measure for Children and Youth (PEM-CY) has shown acceptable psychometric properties, including internal consistency values ranging from 0.67 to 0.80, and test–retest reliability (0.67–0.93, p < 0.0001) in Turkish populations [13]. However, these instruments have limitations. The clinical utility of existing scales in Turkey is constrained by psychometric gaps; notably, the construct validity of the PEM-CY remains unexamined for the Turkish population. Without empirical evidence of its factor structure through confirmatory or exploratory analyses, it is difficult to ensure that the scale accurately captures the intended constructs within this specific cultural context [13].Construct validity of the ECEQ in Turkish children reported acceptable GFI values (≥ 0.90); however, the reported RMSEA values often exceeded the 0.08 threshold, suggesting that its structural validity warrants further investigation in broader samples. Also, the ECEQ is suitable for only children with cerebral palsy [9, 10]. While the HOME focuses primarily on the home environment, it does not capture participation in educational and community contexts [11, 12]. The PEM-CY assess multiple domains of participation and environmental influences; however, their length and complexity may limit feasibility in clinical settings. In contrast, the Environmental Restriction Questionnaire (ERQ) was developed to assess environmental barriers from the parent’s perspective in young children aged 4–6 years, including those with typical development and moderate developmental disabilities. The ERQ is grounded in the International Classification of Functioning, Disability and Health (ICF) framework, which conceptualizes participation as a dynamic interaction between health conditions, personal factors, and environmental influences. For children with developmental disabilities, this framework highlights the lived experience of disability by emphasizing how physical and environmental barriers restrict engagement in meaningful activities and limit social participation [14, 15].

The ERQ categorizes environmental restrictions into three ecological domains—home, education, and community—providing a robust conceptual framework that enhances the scale’s interpretability and psychometric validity [16]. While the original ERQ demonstrated acceptable internal consistency (α = 0.75–0.91) and construct validity [14], its factor structure often varies across different cultural contexts. Such variability, especially when model fit indices are suboptimal, underscores the necessity for rigorous cross-cultural validation. Adapting the ERQ into Turkish is crucial for identifying environmental barriers in local settings, thereby supporting public health strategies aimed at enhancing social participation and long-term outcomes for children with developmental disabilities. Despite the availability of participation scales in Turkey, there is a lack of tools focusing specifically on perceived environmental restrictions in early childhood disability. The ERQ fills this gap by assessing environmental barriers in home, education, and community settings. Consequently, the Turkish adaptation of the ERQ is crucial for supporting clinical interventions and public health strategies designed to improve participation.

The aim of this study is to carry out the cross-cultural adaptation of the Turkish version of the ERQ and to examine its psychometric properties. We hypothesized that, consistent with the original structure, the Turkish ERQ would show adequate reliability and validity.

Methods

Study design

This cross-sectional study was conducted as a psychometric evaluation (reliability, validity, cross-cultural adaptation).

Participants

Sample size adequacy was determined based on established guidelines for factor analysis and scale validation studies. It has been suggested that the sample size should be at least 5–10 times the number of items to ensure stable factor solutions [17, 18]. In addition, for confirmatory factor analysis, a minimum sample size of 200 is generally recommended to obtain reliable and stable parameter estimates [19, 20]. Considering that the ERQ consists of 35 items, the required sample size was estimated to be between 175 and 350 participants. In this study, a total of 270 parents of children with developmental disabilities were included, which meets these recommended criteria and is considered adequate for the planned analyses.

A total of 270 parents of children with developmental disabilities were invited to participate in the study, selected by convenience sampling, which continued throughout the study period. Of these, 50 were excluded (30 due to missing data and 20 for not meeting the age criteria). As a result, the study consisted of 220 volunteer parents of children aged 4–6 years with developmental disabilities. Thirty of these parents completed the test-retest reliability assessment two weeks later. The flow diagram of the study was presented in Fig. 1.Participants were recruited from rehabilitation centers providing education and treatment services for children with developmental disabilities in Turkey. Children and their parents were approached through these centers. The diagnosis of developmental disabilities was confirmed based on official medical reports issued by multidisciplinary hospital committees, including specialists such as pediatric neurologists, child psychiatrists, and other relevant professionals. The inclusion criteria were that the child should be aged 4–6 years with a developmental disability with an official medical special needs report, and be a native Turkish speaker. Participants were excluded if there were missing data or if the child’s primary caregiver could not be reached. This study was conducted at the Bolu Abant İzzet Baysal University. Data were collected between April 2023 and January 2024.

Fig. 1.

Flow diagram of the study

Written informed consent was obtained from all participants and caregivers, indicating their voluntary participation. The study was conducted in accordance with the principles of the Declaration of Helsinki and received ethical approval from the Clinical Research Ethics Committee of of Bolu Abant İzzet Baysal University (Decision number: 2023/45, Date: 21.02.2023).

Measurements

Descriptive information was collected for the children included in the study, including age, height, weight as well as caregiver characteristics such as education level. In addition, the Child and Adolescent Scale of Participation (CASP) was used to assess the validity of the Environmental Restriction Questionnaire (ERQ). All assessments were conducted through face-to-face interviews with the parent in the presence of the child by a qualified physiotherapist in a quiet and relaxed atmosphere.

The Child and Adolescent Scale of Participation (CASP)

The Child and Adolescent Scale of Participation (CASP) is a questionnaire developed based on the ICF-CY to assess the level of participation and associated limitations in children’s daily life situations and activities at home, school, and in the community. It consists of 20 items divided into four subdomains: Home Participation (6 items), Neighborhood and Community Participation (4 items), School Participation (5 items), and Home and Community Living Activities (5 items). Participation and limitations are rated on a scale as follows: “age-expected (full participation), somewhat limited, very limited, unable to participate, and not applicable” [21]. Higher scores indicate better levels of participation. The Turkish version of the CASP has established validity, reliability, and cultural adaptation [22]. The questionnaire was completed by the parents.

Environmental Restriction Questionnaire (ERQ)

The Environmental Restriction Questionnaire (ERQ) is a parent-report measure designed to assess the participation of preschool children aged 4–6 years, both with and without developmental disabilities. It aims to evaluate parents’ perceptions of how environmental factors restrict their children’s participation in daily activities. The ERQ consists of 35 items encompassing a wide range of physical and human environmental factors across home, educational, and community settings. Each item is rated on a 6-point Likert scale, ranging from 1 (“does not restrict at all”) to 6 (“restricts to a very great extent”). Additionally, parents may select the “not applicable” option if a particular item is considered irrelevant, for example, families with no television may mark item 29, “Family television viewing habits,” as not applicable. “Not applicable” responses were scored as zero in accordance with the original ERQ scoring procedure. In line with the original ERQ methodology, “irrelevant” responses were excluded from total and subscale score calculations. However, for psychometric analyses, including confirmatory factor analysis and internal consistency assessment, these responses were coded as zero in order to retain a complete data matrix.

This approach was necessary because methods such as Cronbach’s alpha and confirmatory factor analysis rely on complete item-level data, and the presence of even a single missing observation may lead to listwise deletion of cases, thereby substantially reducing the effective sample size. Given the variability in children’s functional and developmental characteristics, a considerable number of “irrelevant” responses was expected across participants. Therefore, a data completion approach was adopted to preserve the full sample and enable robust psychometric evaluation of all items. The average duration of administration of the scale is approximately 20 min. In the original study, information regarding the ERQ and the CASP was provided to all mothers by trained occupational therapy students, and the ERQ and the CASP was subsequently completed independently by the mothers at their own convenience.

The original ERQ consists of 35 items. Although previous studies have reported a reduced 33-item structure based on factor loadings, the present study was conducted using the full 35-item version of the instrument. No items were removed during the translation and adaptation process in order to preserve the original content and ensure conceptual equivalence. According to the confirmatory factor analysis performed by Rosenberg et al., the 33 items of the ERQ can be grouped into three main domains: Home Restriction (8 items), Community Restriction (12 items), and Education Restriction (13 items). Two items (item 18 and item 27) that loaded less than 0.35 were removed. Higher scores indicate greater participation restrictions. The psychometric evaluation of the ERQ revealed adequate internal consistency, the three domain’s coefficients alpha were calculated as 0.75, 0.87, and 0.79 respectively. Cronbach’s alpha for the ERQ measures ranged between 0.75 and 0.91. The correlations between the Home Restriction, Community Restriction, and Education Restriction factors with the ERQ-total were 0.71, 0.84 and 0.83 respectively. All correlations were significant at the 0,01 level (two-tailed) so support the homogeneity of the ERQ. Convergent and divergent validities were measured between the three domains of the ERQ (Home, Community, and Education) and the CPQ (Child Participation Questionnaire) total scores (Diversity, Intensity, Independence, Child Enjoyment, and Parent Satisfaction). Convergent validity was determined as significant, low to moderate correlations between the Home and the Education Factor with all the CPQ participation measures (except Home factor with Diversity (r=-0,13, p = 0,24)). In contrast, the divergent validity of the CPQ was found insufficient because of very low significant correlations between the Community factor and all the CPQ total participation measures expect Enjoyment [14]. Information regarding the ERQ and the assessment procedures was provided to all mothers by the researchers. The ERQ was administered as a parent-reported questionnaire, and mothers completed the questionnaire through face-to-face interviews at their own convenience. Parents were asked to use their own perceptions to rate the extent to which their child’s environment restricts participation.

Procedure

Permission to translate the original questionnaire into Turkish was obtained from the original author via email. The translation process was carried out in accordance with cross-cultural translation, adaptation, and validation guidelines [23, 24]. The original ERQ was forward translated by two bilingual speakers: one with expertise in the neurodevelopmental disease and the construct being measured, and the other without prior knowledge of the subject matter. Two translations were evaluated by researchers and then combined into a single translated version. Translated version translated back to English by two independent bilingual speakers who were naive in both the neurodevelopmental disease involved and the construct to be measured. These translators’ mother tongue was Turkish, but they were proficient in English. All translation versions were reviewed by a bilingual expert committee involving occupational therapists, physiotherapists, translators, and a biostatistician to resolve discrepancies, ensure cultural and conceptual equivalence, and confirmed that the Turkish version retained the intent of the original instrument. Then the committee decided on the pre-final version. After that, the pre-final Turkish version ERQ was tested for understandability and cultural suitability with 30 parents of children with developmental disabilities. This stage involved cognitive interviewing was conducted to examine participants’ comprehension and interpretation of the items. Feedback obtained from the interviews guided minor revisions to enhance clarity and cultural appropriateness. Upon the expert committee’s approval of the final Turkish version, the adaptation procedure was concluded.

Statistical analyses

The data were analyzed using SPSS 25.0 (IBM SPSS Statistics 25 software, Armonk, NY: IBM Corp.) and AMOS 23.0 software packages. Continuous variables were presented as mean ± standard deviation and median (IQR: 25th–75th percentiles), and categorical variables, as frequency and percentage. Intraclass correlation coefficients (ICC) were calculated to assess test–retest reliability using a two-way mixed-effects model with consistency agreement. Internal consistency of the scale was evaluated using Cronbach’s alpha coefficient. Item reliability was assessed using corrected item–total correlations and Cronbach’s alpha values if items were deleted. Prior to confirmatory factor analysis (CFA), the assumptions for maximum likelihood estimation were evaluated. Data distribution was assessed using skewness and kurtosis values, and no severe deviations from normality were observed. In addition, no significant outliers were detected, and the dataset was considered suitable for CFA using maximum likelihood estimation. CFA was performed using the maximum likelihood estimation method to evaluate the construct validity of the scale. The predefined three-factor structure of the original scale was tested. Standardized factor loadings were examined, and model fit was evaluated using multiple fit indices, including χ²/df, root mean square error of approximation (RMSEA), standardized root mean square residual (SRMR), goodness-of-fit index (GFI), comparative fit index (CFI), adjusted goodness-of-fit index (AGFI), normed fit index (NFI), relative fit index (RFI), and incremental fit index (IFI). Model fit was interpreted based on commonly accepted threshold values. Criterion-related and concurrent validity of the scale were assessed using Spearman correlation analysis. A p-value of < 0.05 was considered statistically significant in all analyses.

Results

The mean age of the participating children was 4.96 years (SD = 0.85). The mean body weight and height of the children were 20.93 ± 4.55 kg and 108.16 ± 12.92 cm, respectively. Of the 220 children, 157 were boys (71.4%) and 63 were girls (28.6%). The most common diagnosis reported were autism (n = 65, 29.5%), cerebral palsy (n = 46, 20.9%), speech disorder (n = 25, 11.4%), and Down syndrome (n = 17, 7.7%). The majority of families reported having a moderate income (n = 176, 80%). Parental education levels were reported as follows: literate (n = 5, 2.3%), primary school (n = 49, 22.3%), middle school (n = 46, 20.9%), high school (n = 78, 35.5%), and university (n = 42, 19.1%). Household income levels were categorized as lower (n = 35, 15.9%), medium (n = 176, 80%), and upper (n = 9, 4.1%) (Table 1). The most frequently reported environmental restrictions in each domain were identified based on item-level mean scores. As presented in Table 2, items with higher mean scores within the home (Factor 1), education (Factor 2), and community (Factor 3) domains were interpreted as representing the most prominent environmental restrictions perceived by parents.

Table 1.

Demographic characteristics of children

Age (year)	Mean ± S.D.	4,96 ± 0,85
	Med (IQR)	5 (4 - 6)
	min - max	4 - 6
Height (cm.)	Mean ± S.D.	108,16 ± 12,92
	Med (IQR)	108,5 (100–118)
	min - max	70–150
Weight (kg.)	Mean ± S.D.	20,93 ± 4,55
	Med (IQR)	20 (19–22)
	min - max	12–42
Body mass index (kg/m2)	Mean ± S.D.	18,14 ± 3,92
	Med (IQR)	17,7 (15,71 − 20)
	min - max	9,92–32,87
		n	%
Gender	Girl	63	28,6
	Boy	157	71,4
Educational level of parent’s	Literacy	5	2,3
	Primary school	49	22,3
	Middle school	46	20,9
	High school	78	35,5
	University	42	19,1
İncome level	Lower	35	15,9
	Mediate	176	80,0
	Upper	9	4,1
Diagnosis	Down syndrome	17	7,7
	Cerebral palsy	46	20,9
	Hydrocephaly	1	,5
	Speech disorder	25	11,4
	Autism	65	29,5
	Attention deficit and hyperactivity	2	,9
	Epilepsy	1	,5
	Spina bifida	12	5,5
	Orthopedic disability	4	1,8
	Hearing disability	7	3,2
	Obstetric Brachial Pleksus Paralysis	1	,5
	Musculer dystrophy	1	,5
	Mental-motor retardation	6	2,7
	Genetic syndrome	12	5,5
	Developmental delay	17	7,7
	Cortical anomaly	3	1,4

n Frequency, % Percent, Mean ± S.D Mean±Standard deviation, Med (IQR) Median (25.-75. Percentiles), min – max minimum – maximum values

Table 2.

Environmental Restriction Questionnaire test-retest and internal consistency analysis results

	ICC	95% C.I. Lower	95% C.I. Upper	Mean ± S.D.	Corrected Item-Total Correlation	Cronbach’s Alpha if Item Deleted
ERQ1	0.742	0.232	0.913	2,65 ± 2,07	0,579	0,934
ERQ2	0.813	0.443	0.937	2,75 ± 2,01	0,503	0,934
ERQ3	0.712	0.141	0.903	3,33 ± 2,16	0,449	0,935
ERQ4	0.612	-0.156	0.87	3,66 ± 2,17	0,405	0,935
ERQ5	0.738	0.221	0.912	3,26 ± 2,1	0,424	0,935
ERQ6	0.8	0.404	0.933	3,27 ± 2,16	0,497	0,934
ERQ7	0.77	0.316	0.923	2,9 ± 2,08	0,567	0,934
ERQ8	0.852	0.558	0.95	2,95 ± 2,06	0,538	0,934
ERQ9	0.725	0.181	0.908	2,45 ± 1,94	0,543	0,934
ERQ10	0.831	0.496	0.943	2,22 ± 1,85	0,591	0,934
ERQ11	0.599	-0.196	0.865	2,1 ± 1,86	0,548	0,934
ERQ12	0.77	0.316	0.923	2,43 ± 1,96	0,57	0,934
ERQ13	0.874	0.624	0.958	2,94 ± 2,18	0,485	0,935
ERQ14	0.843	0.531	0.947	2,46 ± 2,05	0,578	0,934
ERQ15	0.823	0.472	0.94	2,74 ± 2,03	0,521	0,934
ERQ16	0.826	0.483	0.942	2,16 ± 1,76	0,589	0,934
ERQ17	0.675	0.031	0.891	2,32 ± 2,02	0,573	0,934
ERQ18	0.789	0.372	0.929	2,64 ± 2,26	0,511	0,934
ERQ19	0.957	0.873	0.986	2,92 ± 2,14	0,43	0,935
ERQ20	0.888	0.666	0.962	2,56 ± 2,14	0,56	0,934
ERQ21	0.818	0.458	0.939	2,52 ± 2,25	0,581	0,934
ERQ22	0.866	0.6	0.955	2,58 ± 2,26	0,574	0,934
ERQ23	0.886	0.661	0.962	2,65 ± 2,26	0,629	0,933
ERQ24	0.74	0.227	0.913	2,58 ± 2,22	0,613	0,933
ERQ25	0.721	0.168	0.906	2,61 ± 2,29	0,581	0,934
ERQ26	0.931	0.794	0.977	2,86 ± 1,96	0,557	0,934
ERQ27	0.893	0.68	0.964	2,75 ± 2,13	0,513	0,934
ERQ28	0.834	0.506	0.944	3,1 ± 2,32	0,552	0,934
ERQ29	0.9	0.702	0.966	2,7 ± 1,87	0,458	0,935
ERQ30	0.974	0.922	0.991	2,43 ± 1,87	0,607	0,933
ERQ31	1	1	1	2,26 ± 1,73	0,597	0,934
ERQ32	0.914	0.743	0.971	3,17 ± 2,44	0,435	0,935
ERQ33	0.848	0.547	0.949	4,38 ± 2,68	0,312	0,937
ERQ34	0.909	0.729	0.969	2,94 ± 2,18	0,468	0,935
ERQ35	0.986	0.958	0.995	2,36 ± 1,92	0,607	0,933
ERQ_Factor_1	0.809	0.43	0.936	19,77 ± 10,89		0.847
ERQ_Factor_2	0.888	0.665	0.962	36,21 ± 15,76		0.858
ERQ_Factor_3	0.929	0.788	0.976	35,23 ± 18,33		0.884
ERQ_Toplam	0.909	0.728	0.969	96,6 ± 41,32		0.936

ICC Intra Class Correlation coefficient, %95 C.I %95 Confidence Interval

Reliability of the ERQ

The test-retest reliability results demonstrated excellent reliability for the ERQ total score and its subscales, with intraclass correlation coefficients (ICC) ranging from 0.80 to 0.90 (Table 2). The Cronbach’s α values for the ERQ total score and its subscales ranged from 0.84 to 0.93, indicating good internal consistency.

The analysis of internal consistency and reliability revealed that all items demonstrated high internal consistency and reliability, with no items requiring removal or revision (Table 2).

Validity of the ERQ

The 35 items of the ERQ were examined across three sub-dimensions, and the results of the confirmatory factor analysis are shown in Fig. 2. Factor 1, represents the Home Restriction Measure; Factor 2 represents the Education Restriction Measure and Factor 3, represents the Community Restriction Measure.

Fig. 2.

Confirmatory factor analysis of Environmental Restriction Questionnaire

Based on the fit index values obtained from the confirmatory factor analysis, the model demonstrated poor model fit (p < 0.001) (RMSEA = 0.112 / GFI = 0.648) (Table 3). These findings indicate that the proposed three-factor structure was not adequately supported in this sample.

Table 3.

Fit criteria of Environmental Restriction Questionnaire confirmatory factor analysis

	Fit indices
	Stage 1	Stage 2
p	0,0001*	0,0001*
χ2	1842,33	1725,93
df	492	461
χ2 /df	3,745	3,744
RMSEA	0,112	0,112
SRMR	0,122	0,121
CFI	0,665	0,679
GFI	0,648	0,656
AGFI	0,598	0,606
NFI	0,596	0,611
RFI	0,567	0,582
IFI	0,668	0,682

χ² Chi-square value, RMSEA Root Mean Square Error of Approximation, SRMR Standardized Root Mean Square Residual, CFI Comparative Fit Index, GFI Goodness of Fit Index, AGFI Adjusted Goodness of Fit Index, NFI Normed Fit Index, RFI Relative Fit Index, IFI Incremental Fit Index

All 35 items of the ERQ were included in the confirmatory factor analysis. Factor loadings of the items are presented in Table 4. Most items demonstrated moderate to high loadings; however, item 33 (the nanny/housekeeping services in our house) exhibited poor psychometric performance with a factor loading below the 0.30 threshold. This item was identified as a significant contributor to the overall model misfit, suggesting that it may be culturally irrelevant or conceptually distinct for the Turkish population. It was evaluated in an alternative model. This modification did not result in a meaningful improvement in model fit; therefore, the original structure of the scale was retained. Because this study aimed to adapt the ERQ rather than develop a new scale, and we wanted to preserve the original content and conceptual scope of the instrument like previous adaptation study [14]. Furthermore, “not applicable” responses were disproportionately high for item 33, reported by 40% of the sample. This suggests that the item may be conceptually distinct or culturally irrelevant for a significant portion of Turkish families, potentially due to variations in culture and socio-economic status of the Turkish sample.

Table 4.

Confirmatory factor analysis of the ERQ

Stage 1			Stage 2
Item	Factor	Estimate	Item	Factor	Estimate
Item10	F1	0,577	Item10	F1	0,578
Item11	F1	0,657	Item11	F1	0,657
Item12	F1	0,703	Item12	F1	0,703
Item13	F1	0,596	Item13	F1	0,595
Item15	F1	0,657	Item15	F1	0,656
Item16	F1	0,802	Item16	F1	0,802
Item17	F1	0,765	Item17	F1	0,765
Item26	F1	0,44	Item26	F1	0,439
Item1	F2	0,686	Item1	F2	0,685
Item2	F2	0,663	Item2	F2	0,66
Item3	F2	0,549	Item3	F2	0,544
Item4	F2	0,516	Item4	F2	0,51
Item5	F2	0,544	Item5	F2	0,541
Item6	F2	0,636	Item6	F2	0,633
Item7	F2	0,732	Item7	F2	0,739
Item8	F2	0,692	Item8	F2	0,699
Item9	F2	0,713	Item9	F2	0,718
Item31	F2	0,545	Item31	F2	0,544
Item33	F2	0,217
Item35	F2	0,566	Item35	F2	0,563
Item14	F3	0,382	Item14	F3	0,382
Item19	F3	0,31	Item19	F3	0,309
Item20	F3	0,765	Item20	F3	0,765
Item21	F3	0,832	Item21	F3	0,832
Item22	F3	0,829	Item22	F3	0,829
Item23	F3	0,896	Item23	F3	0,896
Item24	F3	0,872	Item24	F3	0,872
Item25	F3	0,762	Item25	F3	0,762
Item28	F3	0,446	Item28	F3	0,446
Item29	F3	0,372	Item29	F3	0,372
Item30	F3	0,453	Item30	F3	0,453
Item32	F3	0,328	Item32	F3	0,328
Item34	F3	0,337	Item34	F3	0,337

Concurrent validity of the ERQ was supported. Both the subscale scores and the total score of the CASP scale were found to be statistically significant and weakly negatively correlated with the corresponding scores of the ERQ (r = − 0.351 to − 0.135; p < 0.05) (Table 5).

Table 5.

Correlations in ERQ and CASP

		ERQ Factor_1	ERQ Factor_2	ERQ Factor_3	ERQ Toplam
CASP-Home Participation	r	-0.326*	-0.188*	-0.242*	-0.276*
	p	0.0001	0.005	0.0001	0.0001
CASP-Neighborhood and Community Participation	r	-0.331*	-0.241*	-0.289*	-0.319*
	p	0.0001	0.0001	0.0001	0.0001
CASP-School Participation	r	-0.304*	-0.178*	-0.279*	-0.296*
	p	0.0001	0.008	0.0001	0.000
CASP-Home and Community Living Activities	r	-0.287*	-0.208*	-0.135*	-0.211*
	p	0.0001	0.002	0.045	0.002
CASP-Total	r	-0.351*	-0.237*	-0.286*	-0.326*
	p	0.0001	0.0001	0.0001	0.0001

*p < 0.05 statistically significant correlation; r: Spearman correlation coefficient

When the ERQ translation process completed eight experts (occupational therapists, physioherapists, translators, and a biostatistician) rated each item on a 4-point scale ranging from 1 to 4. None of the items received a score of 1 or 2, and all items were rated as 3 or 4, indicating that they were considered appropriate in terms of content validity. Furthermore, the item-level Content Validity Index (I-CVI) values were 1.00 for all items, and the scale-level Content Validity Index (S-CVI/Ave) was also at the maximum level ((S-CVI/Ave = 1.00). All items were carefully reviewed during the translation and cultural adaptation process, and were retained in their original form to preserve conceptual equivalence with the original instrument.

“Not applicable” responses were observed across all items. The lowest frequency was found for items 16 and 26, each reported by 6 children (2.7%), while the highest frequency was observed for item 33, reported by 88 children (40%). These responses were scored as zero in accordance with the original ERQ scoring procedure and were included in the analyses.

Discussion

The ERQ was designed to evaluate parental perceptions of environmental restrictions affecting their children’s participation (ages 4–6). It has potential as a practical, easily applicable tool for comprehensively assessing the child’s overall environment. Originally developed in Israel with an English translation, the scale is currently available only in English. The aim of this study was to evaluate the psychometric properties (reliability and validity) and cross-cultural adaptation of Turkish version of the ERQ. Although the ERQ demonstrated good internal consistency and test–retest reliability, the construct validity was weak because of poor model fit in this study. This discrepancy suggests that while the items are consistently measuring a construct, the underlying factor structure may not be adequately represented in this sample. This may be attributed to sample heterogeneity, cultural differences, or potential multidimensionality of the construct. The confirmatory factor analysis results indicated poor model fit, suggesting that the original three-factor structure of the ERQ was not supported by the CFA results in this sample. Several factors may explain the inadequate model fit. First, the heterogeneous nature of the sample, including different types and severity levels of developmental disabilities, may have influenced the factor structure. Second, cultural differences in environmental perceptions and participation patterns may have affected how items were interpreted by parents. Third, some items demonstrated low factor loadings, which may have contributed to the overall poor fit indices. In addition, the use of parent-reported data may introduce subjective bias, as responses rely on caregivers’ perceptions rather than direct observation. Furthermore, “not applicable” responses were scored as zero in accordance with the original ERQ; however, this approach may have influenced score distribution and factor structure. Although some items exhibited sub-optimal statistical indices, we opted to retain the full 35-item structure to preserve content validity. Specifically, Item 33 (the nanny/housekeeping services in our house) was deemed clinically essential for capturing the target construct, and its removal would have resulted in a loss of critical diagnostic information. On the other hand, the decision to retain the original structure was based on theoretical considerations and consistency with the original instrument.

Our study found that the ERQ and all domains had sufficient internal consistency (Cronbach’s alpha > 0.70) and indicated adequate homogeneity. Rosenberg et al. used factor analysis for the English version of the ERQ and found that Cronbach’s alpha for the ERQ measures ranged between 0.75 and 0.91. They found it valid and reliable for young children with and without neurodevelopmental disability [14]. The findings of this study demonstrated high test-retest reliability for the ERQ total score and all its domains.

The correlation between the ERQ and CASP provides evidence of concurrent criterion-related validity. A weak to moderate significant correlation was found between the subdomains of the ERQ and CASP. The negative relationship observed between ERQ and CASP scores is attributed to their differing scoring directions (i.e., higher scores on CASP indicate better participation, whereas higher scores on ERQ reflect greater environmental restrictions). Given that these correlations were established using the CASP, a scale with proven validity and reliability, the findings suggest that the ERQ may capture aspects of environmental restrictions and participation through its subscale and total scores. Chien et al. also found that many items of the ERQ correspond to one or more specific environmental factor categories within the ICF-CY, providing evidence of the scale’s content validity in assessing environmental restrictions [25].

The scale consisted of 3 subscales (Home restriction measure, education restriction measure, community restriction measure) and 35 items, similar to the original. Although the model fit indices in our sample were not at the desired level, we preferred to retain the original 35-item structure. The main reason for this was that our study aimed to adapt the ERQ rather than develop a new scale, and we wanted to preserve the original content and conceptual scope of the instrument. In addition, when we reviewed previous adaptation studies of the ERQ, we observed that the full item set was generally retained and, in some cases, more advanced structural analyses such as CFA were not performed [14]. This suggests that preserving the original structure is not unusual for this instrument. It should also be considered that the ERQ includes a wide range of environmental factors related to participation, and these factors may naturally vary across individuals, contexts and sociocultural factors. For instance, parental roles, access to services, and educational practices in Türkiye may differ from those in other countries, potentially shaping how environmental restrictions are perceived and reported. Therefore, expecting a highly uniform factor structure may not always be realistic. Removing items only based on statistical results might have reduced the content coverage of the scale.

It is important to investigate the cross-cultural adaptation of an instrument in terms of conceptual and item equivalencies, as well as its validity and reliability for use in different cultural contexts [26]. From a cross-cultural perspective, it is important to consider how scale items may be interpreted across different cultural contexts. Following the translation process, an expert panel (n = 8) evaluated all items using a 4-point relevance scale. All items were rated as either 3 or 4, with no items receiving lower scores, indicating a high level of agreement regarding item relevance and clarity. The item-level Content Validity Index (I-CVI) values were 1.00 for all items, and the scale-level Content Validity Index (S-CVI/Ave) was also 1.00, indicating excellent expert agreement. According to established criteria, I-CVI values above 0.78 are considered acceptable, suggesting that all items demonstrated excellent content validity [18]. Although the ERQ was originally developed in a different sociocultural setting, no items were removed during the adaptation process in the present study. This decision was supported by expert evaluations and cognitive interviews, which indicated that all items were conceptually relevant and comprehensible within the Turkish context. The absence of item removal may be attributed to the fact that the ERQ primarily assesses universal environmental constructs, such as home, education, and community participation. However, environmental factors and parental perceptions may vary across cultures; therefore, the findings should be interpreted within the Turkish context. A previous study introduced modifications to the ERQ for use with children aged 2–12 years with developmental disabilities; certain items were deemed unsuitable for the target culture and age group and were revised accordingly [25]. Beyond providing a linguistic adaptation, the present study contributes to the field by demonstrating that environmental restrictions in young children with developmental disabilities may vary across domains depending on contextual and cultural factors. In particular, the finding that restrictions were more prominent in education and community domains highlights the importance of environmental accessibility beyond the home setting. These results extend previous research by providing evidence from a different cultural context and emphasize the need to consider sociocultural and environmental variability when evaluating participation restrictions in early childhood. The higher level of restrictions observed in the education domain was an expected finding, given that educational environments often present structural and systemic barriers for children with developmental disabilities. These may include limited inclusive practices, inadequate physical accessibility, and insufficient individualized support. These findings are consistent with the ICF framework, which conceptualizes participation as the result of interactions between environmental and personal factors. This study does not merely replicate the original instrument in another language but provides additional evidence regarding how environmental restrictions are distributed across domains in a different sociocultural context.

In this study, the five most common constraints experienced by all children in the home domain (Factor 1) reflected those in the Australian study. These are respectively: family income, location of crafts, location of paints and papers, location of toys and games and height of water taps. The five most commonly reported limitations in educational domain were: computers employed by the family, spouse’s job, family/children’s TV watching habits, number of children in child’s preschool and child’s participation in preschool. In the community domain, the five most commonly reported restrictions were: Distance of the preschool from home, Distance of shopping centre from home, Distance of child’s friends’ houses from home, location of the community centre and nanny/housekeeping services at home. Previous studies have compared children with developmental disabilities and typically developing peers [14, 25, 27]. The characteristics of environments may differ across countries. For example, access to educational resources, transportation systems, and community infrastructure may vary between countries such as Australia and Türkiye, potentially influencing parental perceptions of environmental restrictions. In the present study, environmental restrictions were found to be more prominent in the education and community domains. This finding may reflect contextual factors specific to Turkey, such as variability in inclusive education practices, accessibility of educational environments, and availability of community-based services for children with developmental disabilities. Similar patterns have been reported in previous studies conducted in Israeli and Australian populations. where school and community environments were identified as major sources of participation restriction, therefore common policies can be developed globally [7, 25]. Cross-cultural comparisons should be interpreted with caution, and further studies are needed to examine measurement invariance and contextual influences across different countries.

The findings of this study have important implications for research, clinical practice, and policy. From a research perspective, the study provides additional evidence supporting the cross-cultural applicability of the ERQ and highlights the need to further investigate environmental restrictions in diverse populations and settings. From a clinical perspective, the ERQ offers a practical and easy-to-administer tool that can help clinicians identify environmental barriers affecting children’s participation. This may support individualized intervention planning by targeting modifiable environmental factors. From a policy perspective, the findings emphasize the importance of improving accessibility in educational and community environments. Addressing these environmental barriers may contribute to more inclusive systems and enhance participation opportunities for children with developmental disabilities. While the 35-item scale provides comprehensive clinical information, further research is needed to examine measurement invariance and explore alternative factor structures that may better fit diverse populations. There are some limitations in this study. First, the heterogeneous nature of the sample limited the study, as a standardized classification system could not be utilized to describe functional severity across all participants. Second, the decision to retain the original 35-item structure despite poor model fit may also be considered a limitation. The poor model fit observed in this study may be attributed to the heterogeneous nature of the sample. Future research can assess environmental restrictions using the ERQ in more homogeneous samples or disability-specific groups, and its validity and reliability can be examined accordingly [25]. Given the heterogeneous diagnostic composition of the sample, no single standardized motor classification system was used to describe overall disability severity. Additionally, longitudinal studies are recommended to explore how environmental restrictions change over time and how they influence participation outcomes. Further studies may also investigate the effectiveness of interventions targeting environmental barriers identified through the ERQ in improving participation levels. Additionally, coding ‘not applicable’ responses as zero may not fully capture the conceptual distinction between the non-applicability of an item and the genuine absence of a restriction. This approach may have influenced the variance structure and factor loadings in CFA, as well as led to a potential underestimation of total scores. Lastly, the CASP was used to assess the construct validity of the ERQ. The authors recommend that future studies further investigate the validity of the ERQ and its domains using more comprehensive and diverse assessment tools that cover all aspects of the scale.

Conclusion

This study provides evidence for good reliability of the ERQ; however, evidence for structural validity of the ERQ is not adequately supported. The ERQ appears to be a reliable, cross-culturally adapted questionnaire; however, its structural validity remains uncertain and warrants further investigation for assessing environmental restriction in young Turkish children with developmental disabilities. Although there are already several questionnaires with established Turkish validity and reliability that are designed to assess environmental restrictions in early childhood [10, 13], the ERQ has the advantage of being more practical, easier to administer, and a more comprehensible parent-reported instrument. Additionally, it has been proved reliable in assisting the prediction of children’s levels of participation [7]. However, its use should be considered with caution until further evidence on structural validity is established. Further research with larger, more homogeneous samples is needed to clarify the factor structure and enhance the structural validity of the measure before it can be recommended for broad diagnostic or research applications.

Authors’ contributions

Conception and design: SAT; Material preparation, data collection and analysis: SAT, ST, DGK, HŞ; Writing – original draft: SAT, ST, DGK, HŞ; Writing - review & editing: SAT, ST, DGK, HŞ. All authors read and approved the final manuscript.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Data availability

The datasets generated and/or analysed during the current study are not publicly available due to ethical restrictions but are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

Ethical approval was obtained from the Clinical Research Ethics Committee of Bolu Abant İzzet Baysal University (Decision number: 2023/45, Date: 21.02.2023). Written informed consent was obtained from all participants and caregivers, and the study was conducted in accordance with the principles of the Declaration of Helsinki.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.