Development and psychometric evaluation of waste separation beliefs and behaviors scale among female students of medical sciences university based on the extended parallel process model

The primary version of the questionnaire was developed in a preliminary study in EPPM framework (Fig. 1), based on the review of scientific texts [13, 18, 19, 27‐30] obtained by extended search in Google Scholar, Cochrane Library, PsychInfo, Scopus, PubMed, Magiran, SID, and Science Direct.

In this regard, the authors prepared item pool to measure the EPPM perceptual and behavioral variables and perceived barriers. For this purpose, the authors asked for oral opinions of six experts in the fields of health education (N = 4), environmental health (N = 1), and biostatistics (N = 1). The process culminated in designing a scale containing 65 items in 12 constructs.

The first five dimensions of the questionnaire for assessing perceptions included “perceived sensitivity” (eight items), “perceived severity” (ten items), “response efficiency” (four items), “self-efficacy” (seven items), and “perceived barriers” (five items). These items were scored with alternatives of “strongly agree,” “agree,” “undecided,” “disagree,” and “strongly disagree.”

The sixth dimension of the questionnaire, “fear” induced by the news related to waste risks, was assessed based on mood adjectives, including frightened, tense, nervous, anxious, uncomfortable, and nauseous with six items, using alternatives of “extremely,” “very,” “moderately,” “not at all,” and “never.”

The seventh to ninth dimensions were used to assess three responses of individuals in terms of fear control processes. The measurement items of these responses were prepared in three sections, namely “message minimization” (three items) and “feeling manipulated” (four items), with alternatives of “extremely,” “very,” “moderately,” “not at all,” and “never,” and “defensive avoidance” (three items), with alternatives of “always,” “very often,” “sometimes,” “rarely,” and “never.”

The 10th–12th dimensions assessed the three responses of individuals regarding risk control, meaning adaptive responses [18]. The dimensions included “attitude” (four items), with alternatives of “extremely,” “very,” “moderately,” “rarely,” and “never”; “intention” (four items), with alternatives of “since now,” “over the next 2 weeks,” “over the next month,” “over the next 6 months,” and “over the next year”; and “behavior” (six items), with alternatives of “always,” “very often,” “sometimes,” “rarely,” and “never.”

The selected alternatives were scored from 1 to 5, with higher scores indicative of positive beliefs and behaviors in line with waste separation.

Content, face, and construct validity methods were used to determine the validity of the questionnaire.

To assess the reliability, a test-retest was carried out on a group of 30 students. The questionnaires were completed by students over two stages with a 1-week interval; ICC was estimated to evaluate external reliability. Cronbach’s alpha was estimated in order to determine the internal consistency and reliability of the questionnaire [35].

To evaluate test-retest reliability, we computed the intraclass correlation (ICC) coefficient using two-way mixed model along with 95% confidence interval. A coefficient of more than 0.70 was considered as excellent stability [36]. Internal consistency was assessed using Cronbach’s α coefficient. The values between 0.70 and 0.95 were conventionally considered as satisfactory internal consistency [36].

The research was approved by the Ethics Committee of Isfahan University of Medical Sciences with the code of IR.MUI.RESEARCH.REC.1398.165. Notably, the subjects were ensured of the confidentiality terms regarding their personal information, and a written informed consent was obtained prior to the research.

Construct validity was assessed through the use of exploratory factor analysis (EFA) with varimax rotation extracting nine factors from the scale, namely “perceived threat,” “efficiency,” “fear control: reacting against the communicator or message,” “behavior,” “intention,” “attitude,” “fear,” “fear control: defensive avoidance,” and “perceived” accounting for 16.72%, 9.04%, 7.70%, 6.02%, 5.95%, 4.91%, 4.65%, 3.84%, and 3.58% of total variance, respectively. A KMO value of 0.878 and P < 0.05 for Bartlett’s test confirmed the data viability for factorability. Table 1 provides the factor loadings of the nine factors extracted from EFA on the 64 items of the measure.

The reliability was confirmed for the whole questionnaire and each factor after performing factor analysis through Cronbach’s alpha. In this regard, Cronbach’s alphas of the first, second, third, fourth, fifth, sixth, seventh, eighth, and ninth factors were estimated at 0.942, 0.791, 0.897, 0.709, 0.896, 0.733, 0.853, 0.850, and 0.879, respectively. Cronbach’s alpha of the entire tool was further calculated at 0.876, thereby confirming the internal reliability of the tool. On the other hand, the ICC was estimated at 0.675, 0.816, 0.975, 0.608, 0.721, 0.736, 0.566, 0.858, 0.753, and 0.889 for the first, second, third, fourth, fifth, sixth, seventh, eighth, and ninth factors, and the entire tool, respectively. This confirmed the external reliability of the tool (Table 2). Table 2 also shows the distributions and the main results based on the instrument among participants.

The first exploratory factor was “perceived threat,” which included 18 items of perceived sensitivity and perceived severity associated with waste risks. It is often difficult to separate the items in measuring the perceived sensitivity and perceived severity. The integration of these two constructs has been formerly reported [38]. According to previous studies and based on EPPM, people have to perceive the severe risk of accumulation and lack of waste separation [39]. Measuring the perception of risk severity and level of vulnerability can help anticipate and justify preventive behaviors [37]. In addition, the assessment of information and risk associated with environmental behaviors has been proposed as an environmental health literacy construct [40]. As such, appropriate tools must exist to measure this variable in several studies and assess preventive behaviors and risk control, including waste separation [41, 42].

The second extracted factor was “efficiency” with 11 items. This factor was obtained by integrating the items of two dimensions in the primary questionnaire, including the efficiency of waste separation behavior in reducing hazards and self-efficacy, which is understanding one’s own ability to separate all wastes. In addition to assessing the perceived threat in the EPPM framework, it is crucial to estimate efficiency [43] in order to determine its role in predicting such healthy behaviors as waste separation. The next step after perceiving threat is to improve perceived self-efficacy, which is effective in preventive behaviors [43, 44]. A study showed that individuals performed waste separation behavior if they believed that separating waste increased the recycling of valuable products, protected recyclable materials against pollution, and increased their reuse [45]. According to EPPM, it seems that the change in attitude, intention, and behaviors such as waste separation is higher in terms of perceiving the related threat and positively assessing the efficacy of recommended behaviors. In this regard, it is extremely crucial to have standard tools to confirm or reject such type of hypothesis [43, 46].

A construct discovered in this study was “fear,” encompassing six items. Fear of negative consequences motivates people to change unwanted behavior patterns. Behaviors can be successfully changed by use of different techniques of fear-based messages [46]. In a study, some families pointed out the significant benefits obtained from recycling metal objects; however, they did not separate plastic bottles due to the fear of negative consequences associated with their reuse [41]. According to Witte, ineffective fear can prevent behavior change and the promotion of healthy behaviors. Therefore, evaluation of fear and its determinants based on EPPM can contribute to the promotion of effective behaviors [18, 37, 47].

The fourth extracted construct was “reacting against the communicator or message” because this 17-item part of the questionnaire assessed the reactions of students to waste warning messages. According to Witte, individuals make such reactions to control fear [18]. This part of the questionnaire was obtained from integrating the items related to measuring two reactions proposed by Witte, including “message minimization” and “feeling manipulated,” which showed a type of maladaptive reaction to risk messages or the communicator. To control their fear, individuals tend to make the risk messages seem worthless and exaggerated [19, 37, 48]. Cismaru asserted that people react against warning messages when they perceive a high level of threat but believe in their lack of ability to reduce the associated risks [47]. On the other hand, if there is a high perception of threat and low perception of efficacy, they only engage in fear control instead of adopting the correct behavior and reducing risk. Therefore, there must be a balance between threat and efficacy in order to promote behaviors such as waste separation. In this respect, designing tools and measuring such variables can guide efficacy enhancing interventions.

Another construct was “defensive avoidance” which, similar to the fourth factor, is a maladaptive reaction of people to control fear, not danger. However, this construct is different in that people who perceive fear but believe in their inefficient behaviors avoid warning messages in any way possible [19, 37, 48]. In the present research, the developed tools assessed the reaction of students to avoiding the reception of messages associated with the risk of wastes and their separation. As previously reported, people often react in a way to reduce their fear (avoiding reading, listening, or talking about the subject) as a consequence of risk messages. According to EPPM, via suitable tools, we can more focus on the prevention of unfavorable impacts of messages on the audience and engaging them in fear control process and lack of behavioral changes [47].

The sixth extracted factor was “attitude” comprising four items. To design a successful program for recycling, we must improve the attitude of people towards environmental issues to promote proper behaviors in this respect [28]. Fabian pointed out that a positive attitude is one of the strongest constructs that more efficiently explains the intention to conduct recycling behaviors [49]. This scholar further posited that the attitude of students towards proper waste recycling behavior was improved by education, which can affect the behavior of families with regard to intergenerational influences [50]. The first step in designing educational programs on waste recycling is to consider the public attitude [41, 51]. The adaptive threat control reactions of individuals can be assessed through the use of suitable tools in order to measure their attitude, intention, and behavior [18, 43].

In the present tool, one of the extracted constructs was “intention,” comprised of four items. Evaluation of intention can show the success or failure of using EPPM [18, 19, 37, 43, 48] and indicate the threat control reaction in individuals. Moreover, it can properly anticipate the actual behaviors of people [52, 53]. Instead of measuring intention, Saphores et al. evaluated and compared the variable of willingness to recycle electronic waste and reported its association with behavior [54]. On the other hand, Wang et al. measured intention and reported its correlation with waste recycling behavior [16]. Another research showed the necessity of identifying the attitude and intention to conduct tourism promotion behaviors in the environment [55]. In a research, Echegaray and Hansstein observed that recognition and perception of complications associated with electronic waste on environment play a significant role in intention [49]. Nonetheless, the direct association between perceiving threat and intention of conducting a behavior has been formerly hypothesized [19]. Accordingly, it seems crucial to design a proper tool for measuring intention and its determinants and related behaviors.

The eighth six-item construct extracted in the tool was “behavior.” According to Witte in EPPM, risk control-related behaviors indicate one’s adaptive responses [18]. Therefore, waste separation behavior in the present research was an adaptive response to controlling the threat of actual risks caused by lack of management. In other words, the key objective is to specify whether the threat and efficiency components in educational messages have been effective or not. According to this model, educational intervention improves the behaviors of people if risk-decreasing methods and behaviors are emphasized in educational messages as far as fear is concerned [56]. The pro-environmental behavior (PEB) has been deployed as a solution to promoting eco-friendly model through measuring, promoting, and teaching waste separation behaviors [57]. A correct waste separation behavior in youth can be created through increasing their perception and concerns regarding waste topic and simple and practical solutions [24]. However, valid tools are required to assess such behaviors.

The ninth construct was “perceived barriers” with five items. This construct was added to the default model following the suggestion of the team of authors and approval of the expert panel. This construct measures barriers to waste separation from the perspective of students. Given the time-consuming nature of the recycling process, this behavior is less employed by manufacturers due to barriers such as weak infrastructures and compulsory rules for regulating long-term programs [49]. According to Echegaray and Hansstein, perceived barriers could prevent behaviors as an independent variable despite the increase in perceived threats and efficiency in a study [49]. Other preventive factors affecting environment-related behaviors include lack of sufficient space in different places [41] and paucity of suitable conditions to recycle paper [52]. In a study, 47.8% of the subjects were eager to conduct waste separation behaviors if such barriers were removed and separate containers were available for waste recycling [41]. Therefore, it is recommended that this item be measured in addition to the model’s construct when using EPPM.