The Inventory of Parental Representations (IPR)
The IPR comprises two sections - one for the mother and one for the father - each including the same 62 items (Appendix 1). A distinction is made between maternal and paternal representations, even if the items are similar. The answers are based on a Likert scale ranging from 1 (Strongly disagree) to 4 (Strongly agree). The different versions of the IPR are described in Appendix 2: number of items, number of dimensions and the dimensions’ names. Studies conducted in the US do not make any IPR scoring recommendations.
In its original version, the IPR comprised seven dimensions, five of which specifically assessed insecure attachment [18]. The 7-dimensional structure was not found when a new exploratory factor analysis was carried out [19]. A study proposed a revised and shortened version of the IPR comprising 19 items divided into five dimensions. The author relied on a new exploratory factor analysis and interpretation of factor loadings to develop this version without any confirmatory analysis [20]. However, this structural instability is not explained/justified.
Sample
In this study, two sample cohorts were enrolled to cover both parts of the study, namely the qualitative part and the quantitative part. A sample cohort comprising 10 French adolescent volunteers between 13 and 18 years of age was enrolled for the qualitative part (sample 1) and participated in a semi-structured interview. The recruitment process was initiated by posting a notice in a medical establishment. Parents could then put forward their child to participate in the study. The informed consent of parents was obtained. All the interviews were conducted anonymously and no personal identifiable information was collected.
For the quantitative part, adolescents between 13 and 18 years of age were enrolled in French educational establishments so that the questionnaires could be completed within the National Education framework (sample cohort 2). Consequently, the study was presented during classes held in colleges and secondary schools, and the questionnaires were to be completed at home. The adolescents completed attachment assessment questionnaires including the IPR and their parents answered questionnaires focusing on socio-demographic data and lifestyle. All participants were volunteers and their responses were completely anonymous.
Study design – Fig. 1
The study design is described in the Fig. 1. For this study, an Expert Committee was constituted, comprising four confirmed French-speaking child psychiatrists and a medical epidemiologist specialized in questionnaire adaptation. Three of the five Committee members were fluent in English. The whole Expert Committee analyzed the original version, while a smaller Committee, comprising two of the four confirmed French-speaking child psychiatrists and the medical epidemiologist, developed the IPR short version. This Expert Committee worked on the cross-cultural adaptation of the French IPR and devised a short version of the IPR.
Cross-cultural adaptation of the French IPR – qualitative part. This focused on content validity (Fig. 2). Good content validity is an essential property of a measurement scale, as it is an evaluation of the degree to which the content of the scale is relevant with respect to the construct it wants to measure and is recommended by the COSMIN group [25]. One of the components of content validity is face validity, which is how people perceive and comprehend items. The Expert Committee reviewed the French translation of each of the 62 items (box 1, Fig. 2). If any doubt was raised about the face validity of an item, it was discussed with adolescents (box 2, Fig. 2), and then considered by the Expert Committee. An English-speaking psychiatrist was contacted if the exact meaning of some items was unclear. Ten adolescent volunteers (sample 1) were individually interviewed on the face validity of the questionnaire. Ten participants were selected in order to saturate responses during the qualitative analysis and receive varied feedback. All 62 items were discussed and specific questions raised by the Expert Committee were put to them. A qualitative analysis of the responses was carried out. If at least two of the ten adolescents interviewed mentioned a problem with understanding an item, it was discussed again by the Expert Committee (box 3, Fig. 2). Based on these analyses and the attachment theory, the Expert Committee proposed a French cross-cultural adaptation of the IPR, used for the remainder of the study (box 4, Fig. 2).
Study of the metric properties of the adapted French version of the IPR – quantitative part. The participants for the remainder of the study were taken from sample cohort 2 (Fig. 3). Two independent groups were randomized: one for an exploratory approach, further named the development group, constituted from adolescent volunteers from schools and colleges; one for a confirmatory approach, further named the validation group, constituted from school-enrolled adolescents and adopted adolescents.
COSMIN Group recommendations were followed in order to determine the number of subjects required for the psychometric analyses. Each group had to have a minimum number of four subjects by item, i.e., 62x4 = 248 responses for each parent (mother and father) to perform the analyses according to the recommendations of the COSMIN group [25].
Based on the classical test theory, confirmatory factor analyses (CFA) were carried out in the development group on the structures proposed by the developers to explore construct validity (Appendix 1).
Compiling a short version of the IPR – quantitative and qualitative part. In the case of mediocre results in the CFA of the existing structures, the development of a new, stable, content-based, IPR structure was planned according to the following steps in the development group. It was considered that the same structure might not be appropriate for the mother and the father.
Step 1. New dimensions were identified from qualitative work carried out by the smaller committee. The 62 items were grouped into new dimensions consistent with the theory. Items or group of items deemed to be irrelevant for measuring attachment were removed and items which posed significant problems in terms of comprehension were also removed.
Step 2. The floor and ceiling effects of responses to the items were taken into consideration in order to select discriminating items. If at least 80% of the sample size had the maximum or minimum response level for an item, it was discussed by the smaller committee and removed if it was not essential for measuring attachment styles.
Step 3. Psychometric properties were analyzed using the development group data based on the structure proposed in step 1. In order to preserve the structure proposed in step 1, CFA were used iteratively in an exploratory approach. Between each CFA, the model was changed according to the modification indices of the model, reflecting the potential improvement of the model in the event of a change in the structure of the questionnaire and to align the consistency the new structure with the content. The aim was to establish a structure with good fit indices and factor loadings > 0.4. If the presence of an item or a group of items was unnecessary in terms of content, it was removed. If some items did not fit into a dimension or did not present properties acceptable in CFA but were important for assessing attachment in qualitative terms, the smaller committee could decide to retain them.
Analyses based on the Rasch item response model were conducted at the same time. If the Rasch analyses yielded unsatisfactory results, the dimension was discussed and modified again by the smaller committee. This process involved going back and forth between several CFA and Rasch analyses, combined with ongoing discussion focusing on content. The Short IPR obtained at the end of this process was used in the remainder of the study.
Confirmatory analyses of the Short IPR – quantitative part. The confirmatory analyses were performed on the validation group data. Based on the Short IPR structure, a CFA and Rasch analysis were carried out using the same criteria.
Data analysis
In all the statistical analyses, as the items did not have the same positive or negative valence, the scores of certain items were inverted so that all the items followed the same direction in the analyses. A mixed statistical method was used to investigate IPR psychometric properties, based on both the classical test theory and the item response theory.
For analyses based on the classical test theory, two fit indices were used to judge the CFA fit: the Comparative Fit Index (CFI), Tucker-Lewis Index (TLI) and the Root Mean Square Error of Approximation (RMSEA). A CFA was considered to have a reasonable fit with the model if the CFI and TLI values were > 0.90 and if the RMSEA was < 0.08; the fit with the model was deemed good if the CFI and TLI ranged from 0.97 to 1.00 and if the RMSEA was ≤ 0.05 [26, 27]. Items presenting mediocre properties have factor loadings below 0.4 or modification indices that strongly linked the item with several other dimensions.
For the item response analyses, a single-parameter Rasch analysis was carried out. As the items were polytomous, a Rating Scale Model was used for the analysis [28]. Item response theory was used to check that latent trait was well covered by the items, particularly insecure attachment, and that the sequence of item response modalities was respected. Model fit was evaluated via the infit and outfit mean square (MNSQ). Significant MNSQ values less than 0.6 or greater than 1.4 were considered as a misfit to the model [29, 30]. The Person Separation Reliability index, which assesses internal consistency by measuring the discrimination capacity of the different response levels, was considered good if ≥ 0.80 [31]. All the analyses were carried out using R software version 3.5.1.
Ethical consideration
The parents’ informed consent was obtained for sample cohort 1. For sample cohort 2, parental consent was deemed to have been given when the parents completed the relevant questionnaires and all the questionnaires had been collected. Socio-demographic data were collected from parents. Consent was deemed to have been given when the adolescent and his/her parents completed all the questionnaires distributed. All questionnaires in the study were addressed to the parents and not to the adolescents. Consent was obtained from the parents if they responded to the questionnaires. The same was true for the adolescents. The inclusion of adolescents was only effective if all the questionnaires sent to the parents were returned. Indeed, the study was completely anonymous, as agreed with the Ethics Committee who had requested that the people running the study did not have access to any adoption data that could allow identification of the participants. The French Ethics Committee validated this method of collecting consent.
This study is an ancillary study to the French AdoAdo study, the primary objective being to compare attachment in internationally adopted adolescents to that observed in non-adopted adolescents. The transcultural adaptation/validation of a French version was provided for in the protocol which was approved by the Ethics Committee of the CHRU (Regional University Health Centre) in Nancy, France on 20/10/2016, as recommended by the authorities at the start of the study.