Validation of the Questionnaire on Reading Preferences and Habits of Primary School Pupils at Risk of Social Exclusion

: The acquisition of the habit of reading is a fundamental issue for primary school pupils, as it contributes to the integral education of the person and, in some cases, may be essential for pupils to remain in the educational system and to escape social exclusion. In order to carry out research on reading promotion in socially disadvantaged contexts, it is necessary to have an appropriate instrument for these proﬁles. The aim of this work was to adapt and validate a questionnaire on reading habits and preferences that is suitable for pupils in the ﬁrst cycle of primary education who are at risk of social exclusion. To this end, a mixed qualitative and quantitative methodology was used to analyse the content validity and the construct validity, which obtained, as a result, a high correlation between the variables, an appropriate factor analysis, a high overall reliability of the instrument and, in general, an adequate model to correctly explain the data of the proposed structure. We concluded by obtaining a 22-item questionnaire that is divided into ﬁve factors and that will be appropriate for determining the reading habits and tastes of disadvantaged pupils, as well as the inﬂuence of families and schools in the task of encouraging reading.


Introduction
Reading is usually related to culture since it is the main way by which knowledge is acquired [1].It could be defined as a process of interaction between the reader and the book through the integration of new information, which is then decoded and transmitted orally or in writing.This multiple process allows for a series of feelings and expressions to develop, from which new constructions are created [2].
The act of reading brings great benefits at the academic and personal level, and it contributes to the enrichment and maturity of the person [3,4].Mata-Anaya [5] considers that literature offers a profound vision of life, helps to manage emotions, can be a means to obtain answers to problems, helps one to know oneself and the world in general, enriches thinking and develops imagination and creativity, among many other benefits.
Guadamuz-Villalobos [6] argues that reading is an activity that gives meaning to reality and that helps to transmit culture and knowledge.Reading not only provides an intellectual interpretation of a text, but it also, on many occasions, brings feelings and emotions to the surface.
On the other hand, quality education has its foundations in the good development of reading skills since they are fundamental to facing the knowledge society in which we find ourselves [7,8].Sánchez-García [9] agrees that the benefits of reading go beyond the school environment, as reading favours social integration and the integral education of the individual.

Theoretical Framework 2.1. The Habit of Reading
Despite the importance of reading in a person's life, it is not among the favourite activities of students in general.Teachers are faced with the complicated task of transmitting this habit [10].
According to Viramontes et al. [11], a habit is defined as a behaviour that is performed on a regular basis.Therefore, the reading habit involves acquiring the practice of reading frequently.Palacios [12] points out that, in order to refer to the term "reading routine", some caveats must be taken into account, such as the following: do not confuse the word "routine" with repetition or with an action that is carried out automatically or unconsciously; -the reading routine is the final result of a learning process that is reached after the complex and dynamic process of reading.
By following the same author, it can be said that the habit of reading is a voluntary, reflexive, renewing and pleasurable act.
The habit of reading frequently is obtained after a learning process that is designed to achieve the inculcation of this habit within daily activities.It starts with the person's interaction with a book within the family, and it is reinforced and enriched at school [2,13].To achieve the development of this habit, it is crucial that the potential reader is motivated for this encounter, which should be seen as an attractive and voluntary opportunity.Therefore, reading must be understood as a pleasant activity during which the enjoyment is combined with learning and personal growth [14].
This approach to reading, for people who are not familiar with this habit, is achieved through reading encouragement.This is defined as the search for an encounter between the reader and the text, and it is based on motivational activities.It is a planned and evaluable process that is made up of activities with specific objectives [6].

Reading Encouragement
Reading encouragement should be presented as a fun event, and the most appropriate methodology for this purpose should be sought [15].The main challenge is to ensure that the pleasure in reading is maintained, which requires a process of guidance and, at the same time, the promotion of reading autonomy [16].The objectives of reading promotion are to encourage interest in reading, to make it a habit, to make the reader active and critical, to develop creativity and imagination through books, to increase the lexical knowledge and to increase the capacity for retention, reflection and synthesis [2].
It should be noted that emotion plays a role in the development of the reading habit, even from the earliest childhood memories, when a range of feelings towards reading are generated.Good experiences with reading are intrinsic factors that will help this process.On the other hand, since it is a pleasure, it should not be imposed.Reading must be voluntary, and it must be motivated by the desire, or even the need, to practice it [17].

Factors That Influence the Acquisition of the Reading Habit
Family and school interfere in the task of inculcating the habit of reading.The circle that is closest to the pupil has the most direct influence on his/her interest in reading.Therefore, families play a fundamental role in the literacy of their children, in general, and in the achievement of the reading habit, in particular [18].Ahmad et al. [19] also argue that families who spend time reading for pleasure with their children have a significantly positive impact on their children's academic performance and reading achievement.
The ways in which families can be involved in the promotion of reading are diverse and they can include visiting the library, playing games so that their children learn new things, reading to children frequently, participating in shared reading and providing reading materials, among other methods [20].
The educational level of families influences the academic performance in the same way as the socioeconomic level does.As a result, families with higher educational positions often have children with higher levels of reading skills [21].On the other hand, students from families of lower socioeconomic backgrounds usually have lower levels of literacy and reading skills, as well as unfavourable attitudes towards reading and, as a consequence, they have lower academic achievement [22].
By following this line, it is concluded that students from low socioeconomic and educational backgrounds have less access to books and lack parental support in the acquisition of the habit of reading.This makes reading encouragement more difficult in these cases [23].This fact creates important inequalities among students, since those who came from rich reading environments, have advantages at school compared to those who do not [10], especially If we focus on literacy and vocabulary richness upon which new knowledge is built, as well as in terms of the preparation for the acquisition of reading skills and the motivation to read [24].
The responsibility for developing the habit of reading lies with all of the educational agents that are around the pupils, and especially when families do not collaborate and the school becomes the only agent that is responsible for this important educational task, which can change the lives of pupils [25].
Such is the case of students at risk of social exclusion who come from families that live in socially disadvantaged areas.These areas that have high rates of poverty and social exclusion, which are caused by long-term unemployment, the high rate of immigration and the practice of illicit activities that is caused by the precarious economic situation.The characteristics of this population, with high rates of unemployment, family breakdowns, and great diversity of nationalities, translate into high levels of absenteeism, dropout, and school failure in the educational sphere [26].

Methodologies for Encouraging Reading
Moreira-Suasti and Carrión-Mieles [2] affirm that, on most occasions, the methodologies that are used by teachers to encourage reading are not effective and, in many cases, they achieve the opposite effect, and especially when they make use of compulsory reading, summaries, etc.In this way, they have a significant influence on the pupils' lack of interest in reading [27].Therefore, it is essential to update the strategies and to use others that are based on reading inferences, interactions with the texts, and an investigation of the students' previous knowledge and preferences, with the intention of using innovative methodologies that really encourage an interest in reading [15].Alvarez-Ramos et al. [15] also highlight the importance of having teachers who are trained and qualified to encourage this habit, and who must practice and value this activity.
Along the same line, Guadamuz-Villalobos [6] upholds the use of digital images to enhance reading-motivation strategies.Information and communication technologies (ICT) offer a wide range of very useful possibilities to encourage reading encounters [28].Through dynamic products, the interaction between the reader and the book is facilitated, which creates a stronger bond between them.In addition, the reading experience is enriched, and new scenarios that are close to the students' interests are created, which greatly help to promote reading [29,30].
Mushtaq et al. [8] note that ICT have had a significant impact on the reading practices of 21st-century learners.Generation Z spend a lot of time in front of screens, and so they are often attracted to this new reading format, with which they are very familiar.For this reason, ICT can be a motivating tool for potential readers of this generation [31,32].

Justification
Prior to the implementation of a reading-promotion intervention, it is essential to analyse the tastes and habits of the students with regard to reading.As previously mentioned, it is essential that reading-promotion activities are in line with their tastes and preferences [33].
For this purpose, there are questionnaires that are suitable for analysing the reading habits and preferences of primary school students, such as Allan Wigfield's model: the "Mo-tivation for Reading Questionnaire" (MRQ) for primary school students.This model identifies eleven dimensions of reading motivation, and it analyses the differences on grades, genders, and times of measurement.This instrument is composed of 54 items [34,35].
Bautista-Casahuaman and Meléndez-Lozano [36] adapted the MRQ questionnaire by reducing it to 32 items.Chysos [37] designed a scale to measure the motivation of primary school students towards reading the "Self-Efficacy Questionnaire in Literary Reading", which is composed of 31 items.
Reátegui [38], for his part, constructed an instrument to measure children's reading interests, which comprises 36 items that are divided into three dimensions: the home environment, the school environment and the competence as a child reader.Moreover, Romero-Marzo [39] also designed a questionnaire to assess the reading habits of primary school students that provides evidence on the time that they spend reading.
Although all of these instruments are validated and they measure the aspects that are referred to in this study, they are not specially designed for students from socioeconomically disadvantaged backgrounds.Therefore, if the aim was to measure the reading preferences, tastes and habits of pupils at risk of social exclusion in the first cycle of primary education, these instruments would need to be adapted.For example, they ask questions for which it is taken for granted that students have books at home, or that families collaborate in the promotion of the habit of reading, to a greater or lesser extent.However, we know that, in these contexts, there are many homes without books, and there are families who do not have acquired reading skills or who do not tend to value literature or make use of it.
Teachers who work with pupils from a disadvantaged area in the city of Granada (Spain) agreed that some of these tests were too long, and that many of the items were not adapted to their realities.For example, the MRQ questionnaire contains the following item: "It is important for me to see my name on a list of good readers"; normally, these pupils have difficulties with learning to read, which is why this type of list is not available in the classroom.Furthermore, only one of the questionnaires contains items that refer to the use of ICT for reading; however, it does not refer to ICT as a motivational element.
In this sense, this study is pioneering in its adaptation of a questionnaire that is suitable for a context of social vulnerability, and that explores, in greater depth, the motivations for reading, which include ICT.These are known to be very attractive resources for pupils, and they could be used to encourage reading, as mentioned above.
On the basis of this, the objective of this study was to adapt and validate an instrument so that it is suitable for application to pupils who are in the first cycle of primary education and at risk of social exclusion in order to collect their reading preferences and habits, if any, to determine whether the reading-promotion activities that are carried out in schools are of interest to them, and their opinions on the use of ICT for this purpose.

Materials and Methods
In order to achieve the objective and to analyse the data that were obtained in this research, a mixed qualitative and quantitative method was used.
The design of this questionnaire went through a rigorous process that was divided into different phases that are fundamental for the creation of instruments [40]: first, a literature review; then, the establishment of the dimensions of the questionnaire, the formulation of the items, the content validity through expert judgement and, finally, a construct validity and reliability analysis.
The qualitative part was performed by analysing the contents of the suggestions that were made by the six experts who evaluated the first draft of the instrument.For the quantitative analysis of the data from the evaluation of the questionnaire, IBM SPSS for Windows, version 28, was used, and, for the confirmatory factor analysis, AMOS, version 24, was used.The instrument was tested by using different statistics, such as the Kaiser-Meyer-Olkin test of the adequacy of the sampling, and Bartlett's test of sphericity, from which were obtained the chi-square and the significance, the total variance, the rotated factor matrix and the reliability statistic through Cronbach's alpha.The extraction method was also used, through principal component analysis and the rotation method, and by using Varimax with Kaiser normalisation.
This research was divided into four different stages: 1.
A bibliographical review on reading encouragement and the necessity to promote it, especially in disadvantaged contexts; 2.
The preparation of a preliminary questionnaire that is suitable for pupils at socialexclusion risk, and that is based on the selection of the most appropriate items from other validated instruments with regard to the reading habits and preferences of primary school students; 3.
An expert-judgement validation and the reformulation of the instrument on the basis of the results obtained; 4.
A statistical study to establish the reliability and the validity, and to obtain the final instrument.
Following these stages, after the literature review, a suitable and adapted questionnaire was developed for use with primary school students from socially vulnerable backgrounds.From a wide selection of studies that contain questionnaires about the reading habits of primary school students, the most appropriate items were selected within the context of this research.[15,34,35,37,38].To choose the items, the age and the characteristics of the students' families were taken into consideration.
The instrument was subjected to a process of content validation through the judgement of six experts, as is performed at the beginning of this type of operation [41,42], and, as a result, we obtained a database of 47 items (two more than the first questionnaire) that consisted of four response choices that were coded from 0 to 3, which correspond to the following qualitative values: Never, Sometimes, Almost always and Always; or Not at all, A little, Quite a lot and A lot. Afterwards, a nonprobability and intentional sampling was carried out with 68 students who were in the first cycle of primary education (6-9 years old) and who were from two schools that were located in the northern area of the capital city of Granada, which is classified as an area that is in need of social changes by the Consejería de Igualdad y Bienestar Social de la Junta de Andalucía [43]: CEIP Miguel Hernández and CC Escolapios Cartuja.
For the pilot test, informed consent was requested from the parents of the students, and authorisation from the Ethics Committee of the University of Granada.
The structure of the questionnaire after the evaluation of the expert judgement was divided into five dimensions, as is shown in Table 1.

Results
The main experimental results that were obtained and their interpretations are described below, according to the two main phases into which the study was divided.

Validity of the Content
The validity of the content refers to the level of the similarity of the items to the theoretical construct [25].For this purpose, an expert-judgement technique was used, which consisted of several people who judged an instrument, object or educational material, or who give their opinion on a specific point, and which took into account that they have a mastery of the field [28].The evaluation was based on the criteria that are set out in Table 2.

Criterion
Variable Description

Criterion 1 Clarity
The item is easily understood, and its syntax and semantics are adequate.

Criterion 2 Coherence
The item has a logical relationship to the dimension or the indicator that it is measuring.

Criterion 3 Relevance
The item is essential or important, and it must be included.
The strategy that was followed for the expert judgement consisted of two parts.In the first part, each item was judged according to the criteria of clearness, coherence and relevance by using a four-level Likert scale, where the value of 1 corresponds to "strongly disagree", 2 to "disagree", 3 to "agree" and 4 to "strongly agree".The next part was an open response, with comments on each item and on each aspect of the instrument.
The selection of experts was based on their experience in the fields of reading promotion and instrument validation (Table 3).Six experts participated in this expert-judgement evaluation.They were invited to participate by e-mail, whereby they were also sent the evaluation template.The result was the elimination of the irrelevant aspects, and the incorporation of suggestions for improvements.This operation was conducted from 25 February 2021 to 3 March 2021.The responses of each evaluation by the experts are compiled in Table 4, where the average, the standard deviation and the index of agreement are shown.The index of agreement was obtained from the frequency, and it is expressed as the percentage of the agreement of each expert with respect to the criteria of clearness, coherence and relevance of each one of the items.
As for the open-ended feedback, all of the appreciations that were made by the experts in every aspect were considered and were implemented with the aim of improving the instrument through the suggestions provided.These changes were added before the instrument was applied to the participants of this study.

Constructional Validity
Once the content validity had been performed, the next step was to calculate the construct validity.The construct validity is used to assess the degree to which the items measure the construct validity correctly [44].The main statistical tests for its calculation are exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) [45].
The EFA provides information about the distribution of the items along the variable and their adequacy, and so it can be used to explore the set of variables that define the items and their internal structures [46].The CFA is used to confirm the item adequacy of a specific variable [47].It is a complementary step to EFA that indicates whether the distribution of items is relevant to the aspect of which they are part.
The pilot test to refine the scale before the final application was carried out with a sample of 68 students.With regard to the demographic characteristics of the sample, 51.5% were girls and 48.5% were boys, and they were between 6 and 9 years old (M = 7.29, SD = 0.978).
For the exploratory factor analysis, the principal component extraction method was applied with Varimax rotation, and a three-dimensional solution was fixed [48].The data that were obtained in the EFA showed optimal factor loading on almost all of the items, with commonalities higher than 0.50 [49].The only item below was the FAM11 (0.470).At the same time, the calculation of the Kaiser-Meyer-Olkin measure for the adequacy of the sampling showed an adequate value (KMO = 0.683), and Bartlett's test of sphericity showed the significance of the data (χ 2 = 1334.125;df = 666; p-value = 0.000).
On the other hand, the overall reliability of the questionnaire was optimal (α = 0.912), as it was close to 1 [50].By considering the data that were obtained in the EFA and the reliability test, it was decided to remove the FAM11 item.After the removal of this item, the global reliability decreased (α = 0.908), and the measure of the sampling adequacy improved (KMO = 0.700).The scale was reduced from 37 to 36 items.The rotated component matrix showed eleven factors, which explained 71.766% of the cumulative variance.
Nevertheless, a careful study of the composition of the different factors suggested the elimination of certain items because they did not saturate over 0.40, or because some of the factors were explained by only one or two items when three or more were the appropriate numbers.From this analysis, the following items (16) were removed: MO1, MO2, MO3, MO5, MO6, PRE7, PRE8, PRE10, PRE11, PRE12, FAM1, FAM10, FAM11, SCH2, SCH7 and SCH8.The rotated components table (Table 5), after the elimination of these items, provided five factors that explained 63.175% of the total variance (Table 6).With regard to the distribution of the items by dimensions, the first dimension explains 32.765% of the variance and it includes seven items, which refer to the aspect "family reading habit".The second dimension explains 9.083% of the variance and includes the four items that group together parts of the initial dimensions of "Family" and "Preferences", which we decided to combine under the dimension "Students' preferences and reading habits".The third dimension explains 7.966% of the variance and it includes four items that are related to the dimensions "Preferences" and "Motivation", which are combined under the dimension "Value given to reading".The fourth dimension explains 7.047% of the variance and groups the four items that correspond to parts of the initial dimensions of "Preferences" and "School".Therefore, it was decided to combine both dimensions into "Appropriateness of school readings".The fifth dimension explains 6.313% of the variance and it groups the three items that correspond to the initial dimension of "school".
On the basis of the latter analysis, the Kaiser-Meyer-Olkin test (KMO = 0.797) indicated that there was a high correlation between the variables.Barlett's test of sphericity was significant (χ 2 = 670.685;gl = 231; p < 0.001), which indicated that a factor analysis was appropriate.

Reliability Analysis
The reliability indicates the internal consistency of the instrument [51].In other words, the instrument will be reliable if what is being measured is measured correctly.To obtain the reliability of an instrument, different statistical tests are used, such as the Guttman test of two halves, the test-retest method or the well-known Cronbach's alpha, which is the most commonly used in these cases [52].For this reason, it was decided to calculate the reliability of the scale on the basis of the Cronbach's alpha coefficient (α).
The global reliability of the instrument was high (α = 0.898).For each of the different aspects, the reliability was as follows: reading habits in the family or at home (α = 0.847); students' enjoyment of reading and reading habits (α = 0.704); the value given to reading (α = 0.770); cooperative work (α = 0.830); the appropriateness of school readings (α = 0.782); and methodologies for encouraging reading in schools (α = 0.715).If any item was removed, the reliability would decrease, unless DM1 it was removed and it would remain the same (Table 7).

Confirmatory Analysis
To validate the factor structure that was extracted after the exploratory factor analysis, a confirmatory factor analysis was carried out by using the maximum plausibility hypothesis.The model consists of 22 observed variables that are explained by five dimensions that correspond to those that were specified after the exploratory factor analysis [53].The structure of the model and the standardised solution are presented in Figure 1.
Figure 1.SEM model of the questionnaire on reading habits of primary school students at risk of social exclusion.F1-family reading habits; F2-students' preferences and reading habits; F3-value given to reading; F4-appropriateness of school readings; F5-methodologies for reading encouragement in schools.
The chi-square index of the absolute goodness of fit of the model to the data (χ 2 = 326.853;df = 226; p = 0.000) shows the existence of a fit between the observed data and the model.The coefficient χ 2 /df = 1.442 is lower than 5 [54].Furthermore, the comparative fit index (CFI), which measures the improvement in the measurement of the noncentrality of a model, obtains CFI = 0.91, and the normalized fit index (NFI), which measures the proportional reduction in the proper fit function when we move from the null model to the proposed model with NFI = 0.89, has a value close to 0.90, while the RMSEA = 0.079 is less than 0.08.These last three indicators show that the model is adequate to explain the data of the proposed structure.
Finally, a 22-item questionnaire was obtained, which was divided into five factors, as is shown in Table 8.

Discussion
In accordance with this paper and as has been shown with the statements of authors such as Guadamuz-Villalobos [2], Le et al. [15] and Miftachus and Yufiarti [16], among others, it is considered essential to encourage reading in pupils who are at risk of social exclusion in schools, keeping in mind that the family is the main agent of influence for the acquisition of the habit of reading, and that this practice must be instilled through motivation and through pleasure in reading.This will contribute to reducing the disparities that are generated by the socioeconomic factor of the student population, which opens up opportunities for academic success and for the social integration of these students.All of these aspects will facilitate the process of inclusion and will allow for the achievement of quality education, to which all education systems aspire.
By way of outlook, it should be noted that, once the validity and the reliability of the research instrument for analysing the reading habits of pupils who are at risk of social exclusion have been achieved, it can be used in a large and representative sample in the different socially disadvantaged educational centres in order to obtain a diagnosis of the situation before an intervention is needed to encourage reading.
The reliability of the final questionnaire, both dimensionally and globally, is very high.This fact confirms that the steps that were taken for the construction of the instrument were adequate, and that it has a very high internal consistency.

Conclusions
On the basis of these results, and by considering the objectives that were established, the questionnaire was subjected to a rigorous process, which guarantees the validity of its content and provides a solution to the objectives for which it was drawn up.Therefore, this research contributes to deepening the construct of reading promotion in disadvantaged contexts through the elaboration, adaptation and analysis of a questionnaire that was designed to carry out a diagnosis on the interventions, habits and reading preferences of students at risk of social exclusion, and who are just starting to have contact with reading.This questionnaire will be applied anonymously, in person and individually, to each student in the first cycle of primary education in order to explain the contents, and to ensure that the items are understood and are answered correctly.
After a rigorous analysis of the instrument and by considering both the qualitative analysis to assess the content, and the quantitative analysis, and by using different statistics, we obtained a high correlation between the variables, an appropriate factorial analysis, a high global reliability of the instrument and, in general, an adequate model to correctly explain the proposed structure of the data.
With regard to the limitations, it is worth mentioning that this study was applied to a very small sample, and thus the study should be extended in the future in order to achieve greater participation, as well as the inclusion of other schools in similar contexts in the province, in the community and in the country.
This study had a favourable report from the Research Ethics Committee of the University of Granada.The committee certifies that this research "respects the principles established in international and national legislation in the field of biomedicine, biotechnology and bioethics, as well as the rights of personal data protection".
This study also had the authorisation of the families or legal guardians of the students, as well as the consent of the participating educational centres.Funding: This study is derived from the doctoral thesis entitled: "Analysis of reading encouragement in Primary Education students in contexts of preferential attention in the northern area of the city of Granada: ICT as an incentive tool", which was funded by the Ministry of Education, Culture and Sport of Spain, and with the reference FPU18/00676.

Institutional Review Board Statement:
The study was conducted in accordance with the Declaration of Helsinki, and approved by Comisión de Ética de la Universidad de Granada (protocol code 1954/CEIH/2021 and date of approval 03/02/2021).
Informed Consent Statement: Informed consent was obtained from all subjects involved in the study.

Table 1 .
Structure of the questionnaire after the qualitative validation of the experts' judgement.

Table 2 .
Criteria for assessment.

Table 4 .
Mean, standard deviation and agreement index on the basis of frequency.
Note.FAM: family reading habit; PRE: students' preferences towards reading; MO: reading motivation; SCH: school involvement in reading encouragement.