1. Introduction

Spirituality and religion are central characteristics of human experience (Boyatzis, 2003; King and Boyatzis, 2004). A Gallup International Association Poll (1999) of 50,000 adults in 60 countries found that 87% of respondents reported being part being of a religious denomination, 63% indicated that God is highly important in their lives and 75% believed in a personal God, or some kind of spiritual force. Pargament & Magyar-Russel (2005) and Pargament (1997) define religion as a unique phenomenon, both, substantially and functionally. It is defined as a search for significance in ways related to the sacred. Pargament (1999, p. 907), refers to sacred as: “the divine, higher powers and God but to qualities that are closely linked to the divine such as holiness, blessedness, transcendence, omnipotence and infinitude” Religion has been referred as: “The supernatural, the ultimate, the institutional, the creedal, the ritual, the experiential, the ethical, the temperament and the directional” (Pargament, 1999, p.5). On the other hand, spiritually is defined as: “the search for the sacred” (Pargament, 2000; 1997). As with religion, spirituality manifests itself, both, in the individual realm as in the collective.

Mercer (2006, p. 505) establishes the following; “There is a tendency in the literature of children’s religious and spiritual lives to put forward a particular construction of childhood and of spiritual experience as normatively-namely, the picture of the child as innocence, and of childhood spirituality as principally about some ethereal and otherworldly but positive sense of awe and wonder. Many of these perspectives sound very middle class, perhaps very “anglo” in tone […]. When we are attempting to learn about or understand the spiritual lives of children, then we need to listens not only to the angel’s footsteps but also for ways children wrestle with demons (figuratively and existentially encountered)”.

Fowler (1981) developed a faith development theory which proposed that children’s spiritual development is parallel with the cognitive developmental changes described by Piaget and the moral development changes described by Kohlberg. In both Piaget and Kohlberg’s theories, children must learn and adapt to challenges in order to continue their faith development. Failure to do so will manifest in negative consequences. In Piaget’s cognitive development theory (1952), he situated children between the ages of 8 to 12 years old in what he called: “The Concrete Operation Stage”. The stage begins with children developing the ability to conserve quantities in their mind. When children are 9-10 years old, they can think rationally in concrete situations. They achieve the concepts of conservation and reversibility; children at this age also know how to classify objects into groups of similar characteristics and how to put them in order. Kohlberg (1984) assigned children between the ages of 8-12 years old in Stage 2 of the Pre-conventional level. For a child in this level, right or wrong is in accordant to his or her self-interest. Children show very little interest in others. Nevertheless, they do show interest if it suits them. Taking this into consideration, Fowler (1981) classified children from the ages of 8 to 12 years old in what he calls: “The Mythic-Literal Faith”. In this stage, children begin to incorporate the stories or beliefs that symbolize his or her belonging in the community. Beliefs are interpreted literally in the same way as morals and attitudes, as provide by their group. Symbolism is viewed in a one dimensional and literal manner. Storytelling gains importance as a way of giving value and unity to experience. Children in this stage view the world as a fair place where reciprocity is the norm. They have the capacity to tell dramatically and very symbolical stories in an effort to give coherence to their experiences, yet they don’t have the capacity to reflect or conceptualized them. The limitation is an excessive reliance in relationships that are reciprocal. This can make children over controlling or too perfectionist for their age. In addition, they could gain a sense of being a bad child because of the mistreatment from significant others.

It is interesting to mention that there are limited instruments to evaluate children’s perspective of God. Hill and Pargament (2003) list a series of measures of religion and spirituality constructs but they are related to health issues and affinity to God. For instance, the construct of “Closeness to God” is measured with the Spiritual Support Scale (Maton, 1989). For the construct of religion as motivating force, there is the Age Universal I-E Scales (Gorsuch and Venable, 1983). For religious support there is the Religious Support Scale (Krause, 1999) and the Religious Coping Scale (Pargament et al, 2000). For religious and spiritual struggle, there is the Religious Strain Scale (Exline, Yali & Sanderson, 2000), the Intrapersonal Religious Conflict Scale (Trenholm Trent & Compton, 1998) and the Negative Reaction Coping Scale (Zinnbaur, Pargament and Scott, 1999; Zinnbaur et al., 1997). As to measure religiosity/spirituality multidimensions, there is the Multidimensional Measure of Religious/Spirituality (Desrosiers and Miller, 2007). Limited research has been done with minority children, as related to these scales.

Nevertheless, the literature is very conservative (not to say reduced) in mentioning quantitative instruments that measure children’s perceptions toward God. Taking into consideration the aforementioned literature and responding to a limited identification of research studies in the topic, especially with minorities’, this research will try to bring some insight as to how a sample of Puerto Rican children conceive God via the creation of a scale, entitled SEPGC (Scale for Evaluate the Perception of God in Children). In Puerto Rico, we are not able to identify any scale that measures children’s perceptions about God. Thus, this is the main reason why this research is necessary because it provides a pioneer effort to present a scale that could be used to evaluate it.

Hypothesis:

H₁: The SEPGC will be reliable according to Kline’s (2005) criteria of a Cronbach’s Alpha of 0.70 or higher, measuring children’s perceptions about God.

2. Method

Participants: The research was approved by an Institutional Review Board (IRB) in order to protect participants and certify as adequate all research procedures. One hundred children (56% girls and 44% boys) were selected by availability between the ages of 8-12 years old from a private protestant academy in the north zone of Puerto Rico. Availability sample procedure enclosed those children who assent to participate in the research, but also those whose fathers provided approbation and consent to participate in a voluntary base. Children were enrolled in third to sixth grade. Parents reported that 58% of the children were protestant, 35% were Catholics and 7% did not provide an answer. When asked if the family attended religious services during the last month, 14% answered “no” and 86% “yes”.

Investigation design: A non-experimental type design, ex-post facto, of a single group was used. The non experimental investigation tries to analyze the presence of variables in a given amount of time (as for instance, what a person or a child thinks in a determine time period about a specific topic) this design does not allow the manipulation of variables (Fraenkel & Wallen, 2003).

Procedures and Instruments:

Assent Form: The document explains to the participant’s parents the intention of the study and the rights as participants of the research. In addition, it explains the confidentiality and privacy aspects, according to ethical principles required by the American Psychological Association (APA) and the Puerto Rican Psychology Association. We also provided a simple assent form for the child.

Demographic Questionnaire: Socio-demographic questionnaire data sheet contained questions related to the life of the participant, which was used to obtain general socio demographic information. This form examined aspects on age, education and religious denomination membership and adherence, among others.

SEPGC: A list of items was initially obtained via the suggestions of a focal group (focal group includes catholic & protestant clergies, psychologist, sociologist and children from 8-12 years old) for the scale as related to the importance of children’s perception about God. A pool of 60 items was initially contemplated to construct the first version. Those items were evaluated by seven experts in the area of sociology of religion, child psychology and test construction, in order to be included in the final version that will be administered at the academy. Each judge had the opportunity to identify as essential or non essential each one of the items. The content Validity Index (CVI) was calculated for each item according to Lawshe’s procedures (1975). The CVR of the original 60 items of the scale ranges between .285 to 1.00. Items which had below .70 were eliminated. The final scale version to be tested had 40 items. Participants responded on a 4- point Likert scale ranging from " I belief” to "I do not belief”, with higher scores indicating a better perception of God. 

General Procedures. The investigation began by locating the academy and obtaining administration permission. Once the academy was chosen by availability, an appointment with the director was made. The letters and the consents required for the study were drafted. Each of the letters contained the explanation, authorization and the objectives of the investigation. The principal investigator met with the Academy’s director with the purpose of obtaining authorization. The Academy’s director authorized the study. Afterwards, we informed the children parent’s through a consent letter. It included the names of the investigator and graduate research assistants, the study’s objectives, the inclusion criteria, confidentiality right’s and the potential risks of the investigation, among others subjects. It also contained the socio-demographic questionnaire. The information corresponding to the date, the place and the time that the study will take place, was informed to the participants by means of large flyers. On various dates, the investigator and research assistants went to the school to collect the consent form from the children whose parents authorized their participation in the study. On the preselected date the investigators carried out the study by going to the student’s classrooms and administering the assent form and subsequently, the scale to the children whose parents consented. Once the data was obtained, the investigator created a data bank using statistical software SPSS-X version 17.0.

Statistical Analysis. An alpha level of .05 was used for all quantitative data analysis. Descriptive statistics such as: mean, median, frequency and standard deviation were performed to explore distribution of socio-demographic characteristics of the participants. Furthermore, the psychometric properties of the test were analyzed using the following criteria: (a) biserial correlation (r_bis) should be equal or greater than .30; and (b) internal reliability (Cronbach’s alpha) should be equal or greater than .70 (Kline, 2005). Finally, an exploratory factor analysis with Varimax rotation was performed to verify possible constructs that the scale could measure and validity issues.

3. Results

Item Analysis and Reliability

Biserial correlation (r_bis) was used to evaluate item discrimination. Of the 40 items evaluated, only 20 had a r_bis equal or greater than .30. Furthermore, the 20 item scale had a reliability index of 0.91. Table 1 presents the discrimination index of the accepted items.

Factor Analysis and Validity. An exploratory factor analysis following the principal component extraction method was performed on the 20 items. Fourteen items obtain an alpha Cronbach greater or equal to 0.70. This extraction method resulted in one principle component with an Eigen value of 9.16 that explains 46% of the variance and groups all 14 items; all items have a factor loading greater than .30 (refer to Table 1 and 3).

Table 1. Discrimination Index and Factor Loadings of Items Accepted

Afterwards, a Varimax rotation was performed on all items left (14 items) to verify factors. The analysis resulted in 3 factors that can explain 70% of the variance. Table 2 presents the Eigen values and Cronbach’s alpha for each factor while Table 3 groups the items for each factor. Furthermore, the total alpha Cronbach of the SEPGC was 0.91, and excellent index according to Kline (2005).

Table 2. Final Varimax Rotation Sums of Squared Loadings and Reliability Index of Each Factor

The first factor group includes 7 of the 14 items of the scale (50%). This factor was named “How Important God is to Me”. Items in this factor included: God is good; the belief that God helps, belief that God is everywhere, the belief that God is one’s friend, the belief that it is important to pray to God and the belief that God exists. The second factor group includes 5 of 14 items of the scale (36%). It was named “How God Helps Us”. Items in this factor include: God is all mighty, that God protects the family; and that God loves us. The third factor group includes 2 of the 14 items (14%) and was named “The Power of God”. Items in this factor include: God can see everything and God is with the person (see Table 3).

Table 3. Rotated Component Matrix

4. Discussion

The results of our analysis demonstrate that the scale developed is valid and reliable for Puerto Rican children. Moreover, the results permit us to conceptualize the ideas that children have about God in terms of a deep, palpable, and emotional connection with a higher power (God). This connection is based on the ever growing knowledge that children in the 8 to 12 year old age group have about God. This knowledge could be based on the religious and spiritual education taught to them by their family or ideological institutions like church and school. Yet this knowledge is conceptualized by the highly prevalent use of literal symbolism and narrative.

Our instrument detects three main factors: “How Important God is to Me”, “How God Helps Us”, and “The Power of God”. These factors are relevant and consistent with what the literature presents concerning the image of God in children (Pendelton et al., 2002). Furthermore, these factors are also consistent with Mercer’s (2006) view on children’s spirituality and social constructs. The children in our sample present spiritual values that reflect social influence regarding positive and benevolent aspects of God. Nevertheless, they also inform how God can be merciful and forgiving when they are in fault. In general, these three factors are related to the positive views of children regarding how God interacts with them and how God manifest His power in moments of fear, faults and weakness.

5. Limitations

The sample selection represents one of the biggest limitations of the study. Because of its small size and non-representativeness of the population, the results cannot be generalized to all Puerto Rican children from the ages of 8 to 12 years old in private or public schools.

6. Recommendations

For purposes of normalizing and standardizing the scale, a study should be conducted in which the sample is chosen randomly with sufficient participants. It should be interesting to explore the ideas that children from public schools have about God and contrast and compare them. Also, children from a bigger variety of denominations or other religious belief should be studied. As mentioned in the literature, qualitative methods have a lot to offer about children’s spirituality. Supplying participants with art materials and asking them to draw or paint their ideas about God, could produce interesting responses that could be analyzed with qualitative procedures and qualitative software (i.e., Nvivo). A content analysis of this data could complement the knowledge gathered from the quantitative scale.

Finally, a similar study can be carried out with children’s population of other cultures. Therefore, it is highly recommended that SEPGC could be tested with others samples from other cultures. Since this is the first study that creates a valid instrument to assess the perception of God in Puerto Rican children/adolescent, we suggest future studies to standardize the scale. In addition, follow-up research examining the convergent and discriminant validity of the scale needs to be done in a near future. Moreover, the scale could be tested with a larger and more representative Puerto Rican subject population in a near future.