Construction and Validation of the Marital Justice Scale Adaptation of Social Justice Theory in Marriage

The main purpose of this study was to construct and validate a scale for measuring marital justice. A sample of three hundred and four voluntary and unpaid married participants (194 females, 110 males), aged between 20 and 35 years old (29.01 ± 4.44 years), were selected randomly through multi-stage sampling in Isfahan, Iran; the participants included in this sample had preschool child/children, were all in the first decade of marriage, and had at least eight grades of education. All participants were asked to complete the Marital Justice Scale (MJS), the Revised Dyadic Adjustment Scale (R-DAS; Busby, Christensen, Crane, & Larson, 1995), and the Marital Conflict Questionnaire (MCQ; Sanai Zaker, 2000). The exploratory factor analysis extracted two factors labelled ‘Procedural/interactional justice’ (twelve items) and ‘Distributive justice’ (eight items) which accounted for 66.70% of the total variance. The convergent and discriminant validity of the 20-item MJS were supported by an expected pattern of correlations between the scale and the measures of marital quality and marital conflict. All correlation coefficients between the mean scores of the MJS and the scores of the RDAS and the MCQwere statistically significant. The obtained internal consistency was markedly high (Cronbach’s α = .97). The test-retest reliability of the MJS was .87. The results suggest that the MJS is a reliable and valid measure; however, further studies should be carried out in other countries, based on different age groups and socio-economic levels, various developmental stages of family life cycles, diverse cultures and sub-cultures, and according to gender difference so as to validate the MJS.

Relationships consist of interpersonal interactions and exchanges.As used in the field of family studies, a basic premise of the behavioral theory framework is that positive marital behaviors enhance spouses' global feelings toward the marriage whereas negative behaviors diminish positive feelings and cause harm to perceptions of the relationship (Markman, 1981).Since the concept of justice is related to humanitarian and ethical standards that describe how we should act and treat others (e.g., Miller, 2001), and considered as an important positive social behavior (Greenberg & Colquitt, 2005), it may probably enhance spouses' global feelings toward the marriage.
The lack of application of psychologically-based notions of social justice to marriage and family studies is remarkable in light of the positive usage of such notions in other research arenas.The issue of social justice is a dominating theme in our daily lives as it reflects a fundamental concern in most of our interactions with others.Concerns about the value of justice in our social lives go back to ancient moral philosophers such as Plato and Socrates (Rawls, 1971).Justice has clear implications for people's identity (De Cremer & Tyler, 2005).This may not come as a surprise as the nature of people's identity is fundamentally relational (Leary, 2002).As Sedikides and Gregg (2003, p. 110) note "the self operates predominantly within the social world".Furthermore, it is well accepted that the social self of people is developed and constructed by information that they receive through social interaction with others (Leary, 2002;Sedikides & Gregg, 2003).In turn, this relational information defines one's level of identity and goals, and ultimately regulates one's social actions (Carver & Scheier, 1998).As social interactions are the medium through which important others shape the opinion persons have of themselves (Hoelter, 1984) and the way they evaluate or feel about themselves (Leary, 2006), justice thus clearly shapes people's identity, motivation and behavior.De Cremer and van Dijke (2009) have noted that justice reflects important social feedback that people attune to and consequently shapes their identity and motivations.Sedikides and Brewer (2001) have reported that justice can have an impact on the self and identity at the individual (personal goals, and self-enhancement), relational (reliance on interactional goals, role-expectations, and dyadic connections), and collective (reliance on collective goals, and group norms) levels.Justice also affects and activates particular goals, beliefs and values (Lord, Brown, & Freiberg, 1999).By activating goals, the motives of people will energise and consequently direct behavior (Carver, 2001).Justice has the potential to influence people's actions and thus guide the process of selfregulation (Kunda, 1999). De Cremer andvan Dijke (2009) noted that justice has a strong motivational component and has considerable impact on people's behavioral responses.
To study this important social aspect of our interactions with others, the psychological literature has distinguished between different types of justice.Specifically, contemporary justice research distinguishes between distributive justice which denotes justice in effort and rewards (i.e., the fairness of the outcome of and rewards received as a result of one's efforts and also fairness in the assignment of responsibilities and the carrying of burdens), procedural justice (i.e., the fairness of the decision-making procedure) and interactional justice (the attention, dignity, honesty and respect and also the correctness and transparency of the information and explanations one receives during the treatment) (Greenberg & Colquitt, 2005).The importance assigned to these different types of justice clearly emphasizes that information about justice has social consequences.For example, receiving fair or unfair outcomes signals our position towards others and the employment of fair procedures signals that we are valued by the decision-maker.As a result, it follows that justice is an important social regulation tool because it helps shape (a) how our interactions with others evolve and are coordinated, (b) how we can evaluate ourselves in the social world in a meaningful way (i.e., self-esteem), and (c) how our motives and identity are regulated (De Cremer & van Dijke, 2009).Recent research on justice has revealed many insights regarding the social regulation potential of justice.De Cremer and van Dijke (2009) reported that according to the new results, receiving outcomes by means of fair distributions, procedures, and respectful treatments (distributive, procedural and interactional justice) does not only lead to fair or even favorable outcomes (Shapiro & Brett, 2005), but also has consequences that directly implicate one's self-definition and social behavior.
The social justice theory was adapted for organizations (Colquitt, 2001;Niehoff & Moorman, 1993).Roch and Shanock (2006) noted that organizational justice received much attention in the organizational study literature because many important organizational attitudes and behaviors can be directly linked to employees' perceptions of justice.Researchers have found that employee justice perceptions have implications for employee attitudes, such as their perceived organizational support (Rhoades & Eisenberger, 2002), commitment to the organization (McFarlin & Sweeney, 1992), and pay satisfaction (Sweeney & McFarlin, 1993).Justice perceptions have also been found to have implications for employee behaviors such as job performance (Adams, 1965).In the field of organizational studies, distributive justice has been defined as feelings of fairness surrounding the allocation of organizational resources including pay, bonuses, terminations, or any other resources that an organization can provide to employees (Deutsch, 1975).Procedural justice has been defined as feelings of fairness regarding the procedures used in an organization (Thibaut & Walker, 1975).Interactional justice has been described in terms of feelings of fairness regarding how one is treated in an organization, typically by one's supervisor, when procedures are enacted (Bies & Moag, 1986). De Cremer (2005) has reported that when employees perceive procedural justice in the decision-making process of their organization, they show increased commitment and cooperation.Nadi and Golparvar (2011) found significant correlations between distributive justice with procedural justice, procedural justice with organizational identification-the perception of oneness with or belongingness to an organization (Mael & Ashforth, 1992)-and cooperative behaviors with organizational identification.Also, Rankin and Tyler (2009) found that when employees perceive their supervisors as procedurally fair in their decision-making, they show increased compliance with norms and expectations and when they perceive their supervisors' respectful treatment they show more voluntary effort to help the organization.
Due to the following reasons, it could be expected that the social and organizational consequences of justice in the family would be similar to those resulting in organizations.First, the family is defined as an important social organization (Wood, 1996) with interactional subsystems and a hierarchical power structure (Haley, 1971;Minuchin, 1974).Thus, the family can be considered as a kind of organization.Second, such consequences have been reported as characteristic of strong families in the family-strength literature.Family strength is defined as social and psychological characteristics which create a sense of positive family identity, promoting satisfying interaction among family members, and encouraging the development of the individual potential of family members (Otto, 1962).Previous research in the field of family strength reported positive family identity (Stinnett, 1979), motivation, cooperation and goal-orientation (Trivette, Dunst, Deal, Hamer, & Propst, 1990), personal worth for the self and others, positive, clarified and respectful interactions and commitment between the spouses (Schumm, Bollman, Jurich, & Hatch, 2001) as the components that characterize the strength of families.Hence, considering the similarity between the social and organizational consequences of justice and the characteristics of family-strength, justice can be an important element in the family context.Third, there are a number of indirect evidences regarding the impact of justice on marriage.For example, Heaton and Albrecht (1991) found that the fair division of household labour and assets influence marriage preservation.Schwartz' (1994) research included couples who had a mix of equity (fairness) and equality from six large cities in the United States.In his study, having a successful marriage usually meant that spouses shared decision making, responsibility, household labour and financial equity where neither person felt economically dependent on the other.Skogrand, Johnson, Horrocks, and DeFrain (2011), in a qualitative study on self-selected couples who believed they had good marriages, reported that the important elements of financial management in their marriages were trust and communication.Wallerstein and Blakeslee (1995) also found that couples with good marriages felt respected, cherished and loved being together.
As noted earlier, the lack of the application of psychologically-based notions of social justice to marriage and family studies is remarkable in light of the positive usage of such notions in other research arenas.This study aimed at constructing and validating 'The Marital Justice Scale' (MJS) as a measure of a person's perception of the level of justice in his/her spouse's behavior and dyadic life.

Method Participants
According to Kline (1994), analyses such as the factor analysis and item analysis must be taken into consideration in determining the sample size in pie-applications; the number of sample participants must preferably be ten times greater than the number of the items.Kline (1994) pointed out that a 200-person sample would be enough to come up with reliable factors.A sample of three hundred and four people (194 females, 110 males), aged between 20 and 35 years old (29.01 ± 4.44 years) with preschool child/children and in the first decade of marriage (6.05 ± 3.46 years) with at least eight grades of education, were selected randomly through multi-stage sampling from the seven zones of the whole fourteen geographic zones of Isfahan, Iran.The Mean age for men was 32.88 (SD = 4.38) and for women was 30.06 (SD = 3.95).Because of the more convenient availability of married subjects with preschool child/children by means of kindergartens, one kindergarten from each zone of the city was selected.
Subsequently, one class of each kindergarten was randomly selected and the instruments were filled by one parent of each child.All participants were volunteers, anonymous and unpaid.Before the administration of the instruments, the participants received a brief introduction about the nature of the research, ethical requirements for confidentiality and voluntary participation.In order not to be influenced by their spouse, the participants were asked to fill out the scales alone.Only the subjects that completed the instruments were included in the analysis.
Similarly, another sample with one hundred participants (50 females, 50 males), aged between 25 and 32 years old (29.17 ± 4.46 years), selected randomly through multi-stage sampling was considered in order to examine the convergent and discriminant validity of the scale.

Measures
Marital Justice Scale (MJS) -First, based on the social and organizational justice literature (Cropanzano, Prehar, & Chen, 2002;De Cremer, 2005;De Cremer & van Dijke, 2009;Greenberg & Colquitt, 2005;Rankin & Tyler, 2009;Sedikides & Brewer, 2001) and some justice assessment questionnaires and scales such as the 20item Organizational Justice Questionnaire (Colquitt, 2001) and the 18-item Organizational Justice Scale (Niehoff & Moorman, 1993) serving as the indicators of employees' point of view about various types of justice (distributive, procedural and interactional justice) in their organization, the preparatory 50-item version of the MJS was developed and included items about the person's perception of distributive, procedural and interactional justice in his/her spouse's behavior and in his/her dyadic life.Most of the items were written from one spouse's point of view about the justice demonstrated by the other spouse or within the marital relationship.A 5-point Likert scale was used: 1 = never, 2 = rarely, 3 = sometimes, 4 = often and 5 = always.After primary investigations, inappropriate items were removed or revised within three stages: 1-The content validity of the MJS was confirmed by four faculty members specializing in psychology who were knowledgeable about social and organizational justice.2-Before data collection from the study sample, a group of 30 individuals participated in a pilot study to give feedback on the time requirements, wording, and item content of the measure.
Participants were asked whether they encountered any difficulty in understanding any of the items.Based on the pilot study feedback and expert feedback, 32 items were selected and item revisions were made before the administration of the scale to the study sample.3-An exploratory factor analysis was conducted on the remaining items.The factor loadings considered meaningful were ≥ .35(Hair, Anderson, Tantham, & Black, 1998) (1995) developed the 14-item RDAS from the original 32-item Dyadic Adjustment Scale (DAS, Spanier, 1976).
The RDAS was developed to serve as a measure of marital quality in a close relationship and yields total adjustment score and three subscales: dyadic consensus, dyadic satisfaction, and dyadic cohesion (Busby et al., 1995).Dyadic consensus measures the amount of marital agreement between partners.Each item of the dyadic consensus subscale is followed by a Likert-type response scale ranging from 0 (Always disagree) to 5 (Always agree).An example of an item from the 6-item dyadic consensus subscale is "Making major decisions".Dyadic satisfaction measures the tension or discord between partners.Each item of the dyadic satisfaction subscale is followed by a Likert-type response scale ranging from 0 (All the time) to 5 (Never).An example of an item from the 4-item dyadic satisfaction subscale is "How often do you and your partner quarrel?".Dyadic cohesion measures the sharing of pleasant activities (Busby et al., 1995).The 4-item dyadic cohesion subscale is followed by two different Likert-type response scales.Participants use a 5-point Likert scale (0 = never, 4 = every day) for one item of the four items ("Do you and your mate engage in outside interests together?"),and a 6-point Likert scale (0 = never, 5 = more often) for the other three items such as ("Have a stimulating exchange of ideas").The psychometric properties of the RDAS (reasonable construct validity and Cronbach's α from .80 to .90) were confirmed by previous research (Hollist & Miller, 2005).Isanezhad, Ahmadi, Bahrami, Baghban, Farajzadegan, and Etemadi (2012) examined the psychometric properties of the Persian translation of the RDSA in an Iranian population and showed that it had reasonable construct validity and internal consistency (Cronbach's α = .86).The internal consistency coefficients (Cronbach's α) of the RDAS and its subscales namely dyadic consensus, dyadic satisfaction and dyadic cohesion in this study were found to be .90,.83,.88 and .76respectively.
The Marital Conflict Questionnaire (MCQ; Sanai Zaker, 2000) -The 42-item Marital Conflict Questionnaire (MCQ) was used in this study.The MCQ yields a total conflict score and seven subscales: partnership reduction (e.g., "When my spouse has a request from me, I pretend to be too busy with other stuff"), sexual relationship reduction (e.g., "When there is a resentment between my spouse and me, niether of us is inclined to initiate sexual intercourse"), increased emotional reaction (e.g., "When I have a quarrel with my spouse, I shout and insult him/her"), increased support seeking from child/children (e.g., "To gain their consent and support in my favour, I satisfy and give in to my child's/children's irrational demands"), increased individual relationship with family and relatives (e.g., "I am alone in the relationship with my family and my spouse has no relationship with them"), reduction of relationship with wife/husband's family and friends (e.g., "In case of dispute with my spouse, I break off my relationship with his/her family and friends"), and separating of finances from each other (e.g., "Without informing my spouse, I have/open a new saving account") (Sanai Zaker, 2000).Each item is followed by a five Likert-type response scale ranging from 1 (Always) to 5 (Never).A higher score in this scale represents a higher level of marital conflict.The psychometric properties of the MCQ (reasonable construct validity and Cronbach's α from .65 to .81) were confirmed by previous research (Afkhami, Bahrami, & Fatehizade, 2007)

Results
The total score of the MJS and the scores of its subscales are expressed by mean and standard deviation.The reliability of the scale was estimated using the Cronbach's α coefficient (Cronbach, 1970) and test-retest reliability.
The construct validity of the MJS was investigated through the exploratory factor analysis.The principal components analysis, the scree test (Cattell, 1966) and eigenvalues ≥ 1 were used.The Pearson product moment correlation between the MJS, the MCQ and the RDAS was used to evaluate concurrent validity.An exploratory factorial analysis was carried out to identify the underlying factors.This analysis showed two factors (Kaiser-Meyer-Olkin = .971),and an oblique rotation method was employed to facilitate the interpretation.Data analysis was performed using the SPSS/PC+ statistics package (version 16.0).

Validity of the Marital Justice Scale
The scree plot of the exploratory factor analysis of the MJS using the extraction method principal components indicated that two factors with an eigenvalue ≥ 1.5 should be retained, accounting for 66.70% of the total variance (F1 = 60.50% and F2 = 6.20%).Examination of the zero-order correlations among items (not shown here) showed that all items were moderately to highly intercorrelated (average range of correlations = .60-.80).Also, the exploratory factor analysis supported the two-factor structure model of the MJS both for female and male samples accounting for 69.25% of the total variance for females (F1 = 62.83% and F2 = 6.42%) and 61.90% of the total variance for males (F1 = 55.46% and F2 = 6.44%).Table 1 shows the mean and standard deviation of the MJS and its subscales for females, males and the sample as a whole.The oblique rotation final solution is presented in Table 2 where all the items are grouped in the two factors.As shown in Table 2, the items that evaluated procedural and interactional justice are synthesized under the same factor (F1).This factor, which we labelled 'Procedural/interactional justice', was constituted by twelve items and the second factor, labelled 'Distributive justice', by eight items.For procedural/interactional justice, the item values loaded ranged from .59 to .96 and for distributive justice, the item values loaded ranged from .58 to .86.The first factor grouped the items about the fairness of the decision-making procedure (3, 5 and 11) and the attention, dignity, honesty, respect and also correct and transparent information and explanations one receives during the treatment (1, 2, 4, 6, 7, 8, 9, 10 and 12).The second factor was constituted by factors about the fairness of the outcome and the rewards received congruent with one's efforts and also fairness in the assignment of responsibilities and the carrying of burdens (13 to 20).All the items had significant correlations with the total score of the MJS, ranging from .64 to .83(p < .001 in all cases).Tucker's Congruence Coefficient (CC; Lorenzo-Seva, & ten Berge, 2006) supported the similarity of a factor in the female group with a factor in the male group (CC F1 = .98i.e., the similarity of F1 between female and male samples; CC F2 = .87i.e., the similarity of F2 between female and male samples).Lorenzo-Seva and ten Berge ( 2006) suggested a value in the range .85-.94 corresponds to Europe's Journal of Psychology 2013, Vol.9(4), 731-743 doi:10.5964/ejop.v9i4.632 Marital Justice Scale (MJS) a fair similarity, while a value higher than .95implies that the two factors or components compared can be considered equal.Note.F1 = Procedural/interactional justice; F2 = Distributive justice.a By using a standard "forward-backward" translation procedure, each item was translated into English.
The construct validity of the MJS was also examined through calculating the correlation coefficients between the subscales and the total score of the MJS (Table 3).As shown in Table 3, all correlations are significant.A separate group of one hundred participants (50 females, 50 males) was recruited in order to examine the convergent and discriminant validity of the MJS.Table 4 presents the correlations between the MJS, the MCQ and the RDAS.As shown in Table 4, all correlations are significant.These results confirm the convergent and discriminant validity of the MJS.The reliability of the MJS was evaluated in terms of internal consistency reliability and test-retest reliability.Marital Justice Scale (MJS) sample of 60 participants (30 females, 30 males) of the total sample who completed the MJS on two separate occasions with a 2 week interval between the surveys.The coefficients for procedural/interactional justice, distributive justice and the overall scale were found to be .84,.81and.87respectively (p < .001).

Discussion
This study aimed at constructing and validating 'The Marital Justice Scale' (MJS) as a measure of the individual's perception of the level of justice in his/her spouse's behaviors and dyadic life.The psychometric properties obtained for the MJS suggest that this instrument is effective and reliable.The 20 items that compose the scale are grouped into two factors, and the percentage of variance explained by the factors is high (66.70%).The items of the distributive factor are adapted to, and consistent with, Colquitt's (2001) distributive justice (four items) and Niehoff and Moorman's (1993) distributive justice (five items) subscales.The items of procedural/interactional justice are adapted to, and consistent with, Niehoff and Moorman's (1993) procedural (five items) and interactional (eight items) justice subscales.The high and positive significant correlations of the MJS subscales with each other, the total score of the MJS and the RDAS, and also the negative significant correlations of the subscales with the MCQ show the reasonable construct validity of the scale.
As discussed, since the family is defined as an important social organization (Wood, 1996) with interactional subsystems and a hierarchical power structure (Haley, 1971;Minuchin, 1974), the social and organizational consequences of justice could be transferred to the family system.The similarity of the identified characteristics of strong families (Schumm et al., 2001;Stinnett, 1979;Trivette et al., 1990) with the consequences of social (De Cremer & van Dijke, 2009;Greenberg & Colquitt, 2005;Sedikides & Brewer, 2001;Shapiro & Brett, 2005) and organizational (De Cremer, 2005;Nadi & Golparvar, 2011;Rankin & Tyler, 2009) justice and also the confirmation of the psychometric properties of the MJS in this study show that justice can be a possible foundamation for family strength.The results of this study can, in a way, confirm the research carried out by Heaton & Albrecht (1991), Schwartz (1994), Skogrand et al. (2011) and Wallerstein and Blakeslee (1995).Through the results of this study, it can be fairly expected that justice in marriage will lead to consequences similar to those in organizations.It seems in a satisfactory marriage, spouses improve their relationship, communication, feeling of worth, commitment, goal-orientation, cooperation and positive individual, relational and collective identity through behavior based on justice in various aspects of their dyadic life.In the field of organizational studies, researchers have found that employee justice perceptions have implications for a wide variety of employee attitudes, such as the perceived organizational support (Rhoades & Eisenberger, 2002), commitment to the organization (McFarlin & Sweeney, 1992), pay satisfaction (Sweeney & McFarlin, 1993), and behavior such as job performance (Adams, 1965).The results of this study provide initial evidence for the probable manifestation of the implications of justice simiar to that seen in organizations, for spouses' attitudes and behavior.More research is needed to provide a comprehensive theoretical explanation for the adaptation of the social justice theory in marriage and family context.
As shown in Table 2, in this study the items that evaluated procedural justice and interactional justice are synthesized under the same factor.The results of this study are in need of extension and replication.Organizational justice researchers had long debated the distinction between procedural and interactional justice.In the field of organizational studies, several researches have proposed that procedural and interactional justice can be distinguished from one another using social exchange theory (Cropanzano et al., 2002).In particular, procedural justice applies more to the exchange between the individual and the employing organization, whereas interactional justice generally refers to the exchange between the individual and his or her supervisor.In an organization, procedural justice is more closely associated with reactions toward upper management and organizational policies, whereas interactional justice is more closely associated with reactions toward one's supervisor and job performance (Cropanzano et al., 2002).Thus, an explanation for the synthesis of procedural justice and interactional justice in this study could be the difference between the nature of hierarchical structure in dyadic life and in an organization.
The population and size of the sample are the limitations of this study which should be mentioned here.Of course, our work should be considered as an initial approach, and further studies should be carried out in other countries, based on different age groups and socio-economic levels, various developmental stages of family life cycles, diverse cultures and sub-cultures, and also according to gender difference.

Table 1
Means and Standard Deviations for the MJS and its Subscales

Table 2
Principal Components Analysis: Oblique Rotated Solution

Table 3
Pearson's Correlation Coefficients between Subscales of Marital Justice Scale (MJS) With Each Other, and With Total Score of MJS

Table 4
Pearson's Correlation Coefficients Between Subscales of the Marital Justice Scale (MJS) With Revised Dyadic Adjustment Scale (RDAS)and Marital Conflict Questionnaire(MCQ)