Factors affecting consumers’ preferences in purchasing organic vegetables using a theory of planned behavior

Organic food consumption has increased, and health and environmental concerns are becoming particularly crucial for policymakers and individual consumers. Meanwhile, consumers are increasingly demanding sustainable practices for organic food production. This research aims to investigate the impact of sustainability and health values on consumer purchase intention and behaviour in selecting organic vegetables. In so doing, consumers of organic vegetables in East Java, Indonesia were chosen as respondents in this research by using the development of a theory of planned behavior. A structured questionnaire presented in Google form was used to survey 116 research respondents online, and the data was analyzed using Partial Least Squares - Structural Equation Modeling. The respondents’ determination has been adjusted to the minimum requirement for the amount of data used in the analysis by taking into account the minimum R-squares and power statistics. The results indicate that the value of sustainability and health effects on purchase intention through mediating consumer attitudes. Attitudes, subjective norms, and perceived behavioural control, on the other hand, influence purchasing behaviour by negotiating purchase intention. The findings of this research are expected to strengthen the development of marketing strategies for organically processed products, especially vegetables, to gain potential consumers.


Introduction
Consumer concerns about health, well being, and the environment have increased the consumption of organic food in recent years [1].Meanwhile, health and environmental concerns are increasingly important for policymakers and consumers in the food market [2].Consumers demand sustainable food increasingly and are willing to participate in food sustainability processes, and higher levels of consumer education mean greater sustainability concerns [3].
The theory of planned behaviour has been successful in explaining consumer intentions and behaviour [4].Meanwhile, the use of food eating values and individual exploration of purchasing behaviour characteristics is a development of planned behaviour theory in predicting attitudes and intentions toward organic food consumers [5].Consumer behavioural control can be enhanced by improving the labelling system to strengthen the purposes of buying organic products to include more actionable information, such as the health, environmental and social benefit of products [1].
Consumers are generally highly familiar with the practice of organic vegetables, although their subjective knowledge is moderate to low.A key factor in explaining the use of organic plants was subjective knowledge [6].Europe-wide research finds that the consumption of vegetables in 1306 (2024) 012028 IOP Publishing doi:10.1088/1755-1315/1306/1/012028 2 adolescents is low.Therefore, strategies should improve general healthy eating patterns and increase vegetable preference in teenagers [7].
There is a link between subjective norms, perceived behavioural control and environmental concern in different contexts with various purchasing intentions [8].Perceived behavioural purpose and control influence the purchasing behaviour to have the most significant impact on behaviour.Attitudes influence consumer intentions to buy organic vegetables most effectively, followed by perceived behavioural control and subjective norms [9].
Consumer behaviour is a key element for the transition to organic products, and there are different factors affecting consumer preferences, planned, impulsive and unforeseen buying choices [10].The nutritional value, sensory quality and overall health consequences of food choices are major drivers of food choice.Meanwhile, organic and environmentally unfriendly attitudes to genetically modified foods have become correlated with decreased purchase behaviour, suggesting the connection between attitudes towards specific methods of food manufacture to allow real purchases to be predicted [11].
Product attributes that emphasize the environmental impact of production and the social responsibility of food producers have been shown to be valued by consumers who are environmentally conscious and concerned about producing goods.Consumers who are health-conscious about their lifestyle and diet have been shown to value product nutrition information highly [12].Attitudes and health awareness were better predictors of intent to purchase organic food.Consumer awareness moderates the intention to buy organic food positively.
Consumer innovation refers to the proclivity to acquire and use new products more quickly and frequently than others, reflecting a vital personality trait [13].According to the study's findings, millennials are more concerned with sustainability and environmental issues [3].Meanwhile, social values heavily influence consumer ecological concerns [14], and it has been discovered that there is a link between health problems and environmental sustainability in food choices [2].
Previous research suggests that organic food producers and marketers want to keep positive changes in consumer behaviour [15].To expand the organic food market rapidly, policymakers must educate and positively promote organically produced food to consumers, and managers and practitioners can use it to target this niche market and promote organic food in key markets [16] [17].
This research aims to investigate the impact of sustainability and health values on consumer purchase intention and behaviour in selecting organic vegetables.As a result, the findings of this study are expected to strengthen the development of marketing strategies for organically processed products, particularly vegetables, to attract new customers and contribute to the study of consumer behaviour, particularly the development of the theory of planned behaviour.

Research Method
The research is being conducted in East Java, Indonesia, and the method of data collection is nonprobability sampling, more specifically, purposive sampling based on specific criteria [18].Consumers who have consumed organic leaf vegetables such as mustard greens, spinach, caisim, kale, Chinese cabbage or pak choi, lettuce, kai-lan, and others are eligible to respond.The data collection method was an online survey using questionnaires created in Google Form.The online questionnaire was distributed through the WhatsApp social media platform to members of the organic community and organic vegetable producers, who then distributed it to their respective community members and consumers.The procedure was followed to ensure that a representative sample of respondents was obtained in accordance with the research.The number of research samples to be used by considering the significance level and the maximum value of the arrow leads to the constructed variable in PLS-SEM with a statistical power of 80% [19].Thus, at a 5% significance level and a minimum R2 between 0.10 and 0.25, the number of samples suggested is 116 respondents.
While the SEM-PLS analysis responds to the research goals by assessing the construct validity to determine the strong association of the constructs and reliability to measure the consistency and accuracy of the respondents in answering research tools questions [20], convergent validity and discriminatory validity can be demonstrated to measure constructed validity [21].Furthermore, the significance of the relationship between constructs/variables is assessed using the path coefficient, which describes the strength of the relationship between constructs, namely the t-test or critical ratio, R-square value, and validation of the overall structural model with the goodness of fit.Meanwhile, as shown in Figure 1, several empirical studies were used to develop the research model, which discovered that health and sustainability motives related to environmental and social factors influenced consumer attitudes toward organic food [22] [23] [24].Meanwhile, in the theory of planned behaviour model, health and environmental awareness should be included as background factors, and their impact on intention is expected to be mediated by attitudes, subjective norms, and perceived behavioural control [9].On the other hand, the theory of planned behaviour has three concepts: attitude development, perceived control of the behaviour and subjective norms.An individual's intention to perform a particular behaviour is determined by an individual's attitude (i.e., attitude towards purchasing organic foods) [25].Variables are quantified using a Likert scale ranging from 1 to 5 (strongly disagree to strongly agree).The following assumptions and models are therefore suggested: H1 : Attitude has a positive and significant effect on Intention H2 : Perceived Behavioral Control has a positive and significant effect on Intention H3 : Subjective Norms has a positive and significant effect on Attitude H4 : Attitude mediates health consciousnessintention relationship H5 : Attitude mediates sustainability motiveintention relationship H6 : Attitude mediates subjective normsintention relationship H7 : Intention mediates perceived behavioural controlbehaviour relationship H8 : Intention mediates attitudebehaviour relationship H9 : Intention mediates subjective normsbehaviour relationship

Results and Discussions
Table 2 shows 68.97% female and 31.03%male of different ages, educational levels and monthly income, based on 116 respondents collected.This data aligns with previous research in which women dominate organic foods compared to men [26].The fact that men's food is less healthy is the stereotype as well.They favour meat products compared to the more nutritious "female food" as they are more willing to eat vegetables and fruit [27].The largest proportion in this study is respondents 18-20 years of age (40.52 per cent), followed by 21-24 years of age (36.21 per cent).
Moreover, the last level of education for high school equivalent is 50.86 per cent, with monthly earnings below IDR 2,000,000 for the 59.48 per cent of the respondents.Those who completed the online survey showed that the young group (Generation Z) who receive the latest vocational education and a monthly income of less than IDR 2,000,000 was dominant.Similar observations state that the younger population is more likely to purchase organic produce [28] [26].Young consumers are concerned about current environmental problems and positively position the future purchase of green products [29].This research uses reflective model indicators to determine the construct validity test of the research instrument using convergent and discriminatory validity analysis.The dependability of research equipment can be determined during testing by analyzing Cronbach's Alpha and Composite Reliability.Convergent validity testing using reflecting model indicators can be seen by examining each variable indicator's value of loading factors.A rule of thumb with the value of loading factors must have a value of 0.7 or above with a significant p-value to fulfil the criterion of convergent validity [19].Meanwhile, discriminant validity testing can be performed using the Fornell-Larcker approach by examining the value of Correlations Among Latent Variables with Square Roots of AVEs with the terms and criteria that the square root of each AVE construct must be greater than the correlation with other constructs [19] [30].Tables 3 and 4 show the results of the convergent validity and discriminatory validity tests.Furthermore, the value of composite reliability and the value of Cronbach's alpha can be used to test the reliability of research instruments.To meet the general requirements and criteria for reliability testing, the value of composite reliability must be greater than 0.7, or a value of 0.60 to 0.70 can be accepted in exploratory research.In contrast, a value of 0.70 to 0.90 can be considered satisfactory in further study.A value greater than 0.90 (and indeed greater than 0.95), on the other hand, is undesirable because it indicates that all indicator variables measure the same phenomenon, implying that it is unlikely to be a valid construct measure.While other general terms and criteria have a value range from 0 to 1, Cronbach's alpha has a value range from 0 to 1, and if the value is greater than 0.6, it can be said to be reliable [19] [30].5, all variables in the study have a composite reliability value greater than 0.7, indicating that all variables meet the general requirements and criteria for reliability testing.Although the attitude, intention, and behaviour variables have values that are sequentially 0.921, 0.924, and 0.948 close to the value > 0.95, which is an undesirable value, they still meet the reliability testing criteria based on the composite reliability value of more than 0.7 and less than the value unwanted is more than 0.950.While testing using Cronbach's alpha values, all research variables have a value greater than 0.6, indicating that all variables used are reliable using Cronbach's alpha values or that the instrument used in the research is quite reliable.The next step is evaluating the path coefficient for the relation of variables or the hypothesized relationships among structures.Figure 2 shows that health awareness (HC) toward attitude (AT) has a favourable path coefficient value of 0.21 with a p-value <0.01, which means that health awareness has a high positive and significant effect on attitudes.The sustainability motive (SM) toward attitudes (AT) has a path coefficient value of 0.33 with a pvalue <0.01, which means that sustainability motive has a positive and high significance effect on attitudes.Subjective norm (SN) on attitude (AT) has a path coefficient value of 0.33 with a p-value <0.01, which means that subjective norms have a positive and high significance influence on attitudes.Meanwhile, the subjective norm (SN) on intention (AT) has a positive path coefficient value of 0.32 with a p-value <0.01, which means that the subjective norm has a positive and high significance influence on attitude.
In comparison, perceived behavioural control (PBC) has a positive coefficient value of 0.29 with a p-value <0.01, which means that perceived behavioural control has a high positive and significant influence on intention.Attitude (AT) towards intention (IN) has a path coefficient value of 0.30 with a p-value <0.01, which means that attitude has a positive and high significance effect on intention.And intention (IN) on behaviour (BH) has a positive coefficient value of 0.76 with a p-value <0.01, which means that intention (IN) has a positive and high significance effect on behaviour (BH).Moreover, attitude and intention are mediating variables with two segments and three segments.
Figure 2 also shows that the R-Square value is 0.55 or 55 percent for attitudes.Health awareness, sustainability motive and subjective norms are influences on attitude.While the R-Square value for the intention variable is 0.58 or 58 percent, subjective norms, attitudes and perceived behavioural control are affected by perceived behavioral control.Furthermore, the R-Square value for behaviour is 57%, then behavior is influenced by intention.Then the Qsquared on attitude, intention and behaviour has a value of 0.553, 0.585 and 0.572, respectively, thus the results show good predictive validity.
Furthermore, the Full Collinearity VIF evaluation shows the level of collinearity between the predictor variables in order to know whether or not a collinearity problem is indicated.The value of collinearity between variables of health awareness, sustainability motive, the subjective norm of attitude has a value of 2.245, 2.060, and 2.423, which means that it meets the requirements and evaluation criteria for Full Collinearity.Then the subjective norm, attitude and perceived behavioural control variables on intention respectively have a value of 2,423, 2,593, and 1,743, which means they meet the requirements and evaluation criteria for Full Collinearity VIF [19] [30].Likewise, the intention to behavior variable has a value of 2.772 which means that it meets the requirements and evaluation criteria for Full Collinearity.
Evaluation of the effect size value was carried out to determine the size of the influence between variables in the study.The exogenous latent variables of health awareness (HC), sustainability motive (SM), and subjective norm (SN) have effect size values on attitudes (AT) respectively 0.133, 0.215, and 0.199.Thus, it can be said that health awareness (HC), sustainability motive (SM), and behavior control (PBC) have a moderate direct influence on attitude (AT).While the exogenous latent variable subjective norm (SN), attitude (AT) and behavioral control (PBC) have an effect size value on the latent variable intention (IN) respectively 0.254, 0.192, and 0.174 where subjective norms, attitudes and behavioural control have an influence.Moderately direct to intention.And the latent variable of intention (IN) has an effect size value on the latent variable of behaviour (BH) of 0.571, so it has a direct and large effect.
In this research, the Average Path Coefficient (APC), Average R-squared (ARS), Average Adjusted R-Squared (AARS) were 0.362, 0.556, and 0.557, respectively with p-value for all of them p <0.005, so it can be concluded that it is said that all three are quite good and meet the predetermined fit criteria.The value of Average Block VIF (AVIF), Average Full Collinearity VIF (AFVIF) has a value of 1.8 and 2.423 so that it can be said to be ideal because it meets the predetermined fit criteria.Meanwhile, the Tenenhaus GoF (GoF) value has a value of 0.747 which means large or large because the value is 0.36.Furthermore, the values for Sympson's Paradox Ratio (SPR), R-Squared Contribution Ratio (RSCR), Statistical Suppression Ratio (SSR), and Nonlinear Bivariate Causality Direction Ratio (NLBCDR) all have a value of 1 which means ideal because of the fit criteria of the values.It is declared ideal if it has a value of 1 for SPR and RSCR while ideal if it is 0.7 for SSR and NLBCDR.Thus, it can be concluded that all Fit and Quality Indices Models are in accordance with the predetermined fit criteria or rule of thumb, and the models contained in the study can be used.Attitude refers to consumer assessments of some behaviours that can be used as a product-product assessment interaction to predict actual consumer behaviour [31].Attitudes affect organic food procurement positively and significantly [32].Attitude is an important factor that affects purchasing intentions, although each person's attitude differs widely in terms of their preferences.When consumer attitudes and consumer assessments are in line with expectations, the intention to buy environmentally friendly products can occur and can be felt [33].This trend is also supported by research that shows the variance in the interest of consumers to buy organic vegetables and attitudes that affect the intent to purchase and consume organic food together with subjective standards and perceived behavioural controls [9].
Meanwhile, perceived behaviour control reflects a person's perception that resources or chances are available for conduct [34].Perceived behavioral control refers to the perception of a particular behaviour's ease or difficulty.It is also motivated by the availability of opportunities and resources (for example, time, money, skills, cooperation with others).Because these are factors that control actual behaviour, and based on the research results, respondents/consumers have a sense of trust and can take time, resources and willingness to purchase organic vegetables.This is also consistent with the research results into the characteristics of the respondents, in which all groups can reach the products offered based on occupation and level of income.Based on the study results, control behaviour has a higher effect than other variables and is also of greater effect size than other theories of planned behaviour variables.Behaviour controls are more influential because they have a higher effect value than other latent variables because they are excellent indicators or items [35].
Subjective norms have a vital role to play in the choice of organic vegetables by consumers.Subjective norms are a person's perception of social pressure to engage in or not to engage in behaviour, in this case organic vegetable consumers.In some cultures, organic food consumption has more social pressure than in others, which is obvious in the relevance of subjective standards [9].Subjective norms related to health motivation have been shown to be important, while environmental effects have not contributed significantly to predicting the purchases of organic fruit and vegetables [36].Subjective norms, attitudes and perceived behavioural controls can also explain 63% of the difference in consumer interest buying interest in organic vegetables [9].
The awareness of the health of consumers in organic vegetables has a powerful influence.To encourage increased demand for organic products, most consumers will choose food safe for health and environmental safety [37].Inorganic products, vegetables are one of the consumers' most popular organic products after rice, making them one of the main food requirements [38].In the meantime, respondents who bought organic food have a greater health perception, which considers organic food as more nutritious and organic food capable of enhancing resilience [39].Young consumers studied in India also have concerns about current environmental issues and have positive attitudes towards IOP Publishing doi:10.1088/1755-1315/1306/1/01202811 purchasing green products for their future use [29].Then, consumer attitudes towards health and the environment were significantly and positively correlated with organic food, where the attitude component related to health motivation was shown to be significant in predicting the purchase of organic fruits and vegetables [36].
Motive for sustainability strongly impacts the attitudes of consumers when choosing organic plants.A healthy and environmental lifestyle has become a new trend, leaving the old way of living with nonnatural chemicals such as fertilizers, synthetic chemical pesticides and agricultural growing hormones.This healthy lifestyle requires?, safe and environmentally friendliness in food products.Agricultural products are safe for consumption.This highly nutritious and healthy food can be produced using organic farming methods [40].In the meantime, those who do not ever purchase organic foods have an even higher perception that organic farming is good for environmental sustainability and safe for the livelihood of animals and plants [37].
Furthermore, consumers of organic products are still restricted to people with great concern about the sustainability of the environment and health.This concern encourages them to be willing to give organic products premium prices [41].Consumers with higher health and sustainability perceived more favourable positions to purchase organic vegetables based on earlier research [9].
Subjective norms had a very good and important impact on organic vegetables' purchase intentions.Subjective norms shape individual conduct based on social factors or social pressure to exercise or not to exercise certain conducts [41].Besides attitudes that can influence the intentions of purchases or consumer choices, subjective norms also influence attitudes thus, society can influence their decisions [32].The influence or opinion of others can affect a consumer's intention to buy or decide.The exogenous latent variable of subjective norms has been studied due to the influence of the family on certain actions or behaviour.Subjective norms are thus influenced by the intention to purchase organic vegetables, particularly the effects of families.Subjective norms have a positive and significant impact on the sustainability of consumer intentions and the impact of important individuals around them [31].Subjective norms can influence or exert pressure on others to influence certain behavior of an individual to meet the views or expectations of others.Thus the purchase references in subjective norms play an important role.Factors other than family, friends, or others that are considered important greatly impact a person's intention to buy a product.Although studies show that Theory of Planned Behavior (TPB) buildings have the most decisive intention in food shopping and eating, there have also been other studies that find the most influential subjective norms [9].
The intention is the best predictor for real behaviour.The intention is to know the factors affecting behaviour, indicating how hard people are willing to test and how much effort they plan to conduct themselves.The purchase intention is measured with four indicators, namely, a plan to buy, a purchase budget and a tendency to purchase [42].This measurement indicates that the intention to purchase has a major impact on behavioural choices made by anyone.The more precise and real the actual behaviour of the individual is, the higher the intention to perform some behaviours.A positive attitude to food is vital for the taste of vegetables and the recognition of their perceived health benefits that lead ultimately to the intention to eat this food [43].

Conclusions
Overall, this study found that there was a significant effect of health awareness (HC), sustainability motive (SM), and subjective norms (SN) on consumer attitudes in buying organic vegetables (AT).Significant effect of perceived behavioural control (PBC), attitude (AT), and subjective norm (SN) on intention (IN), and considerable influence of intention (IN) on behavior (BH) of buying organic vegetables.The attitude and intention of these two and three segments have been demonstrated to be mediating variables.Organic vegetable business owners should take the opportunity to increase communication about the benefits of buying organic vegetables on health and environmental sustainability, such as in physical stores or social media.Organic vegetable business owners can also interact directly with consumers by providing samples of organic vegetables to increase purchase

Table 1 .
Measurement of the construct to regularly purchase organic vegetables.

Tabel 2 .
The profile of respondents.

Table 3 .
Loading Factors-Based Convergent Validity Test.

Table 3
demonstrates that all indicators in the research instrument have a value greater than 0.5, indicating that they fulfil the overall standards and criteria.This data indicates that behavioural factors can explain the indicators more than 90%, but subjective norms, control behaviour, intentions, health awareness, and sustainability motives may explain the indicators 50% to 80%.As a result, all research indicators are declared valid for measuring the constructs produced, or all variables are reasonably excellent at reflecting their respective indicators.

Table 4 .
Discriminant Validity Test based on Correlations Among Latent Variables with Square Roots of AVEs.Based on Table4, the values in brackets are compared with other values in one row to make the data easier to read and interpret.Thus, the variables of health awareness, sustainability motives, subjective attitudes and norms, purchasing decisions, perceived behavioural control, intentions, and behaviour have AVE square root values greater than the correlation value between latent variables and other variables, indicating that the research instrument has good discriminant validity based on the Fornell-Larcker criterion.

Table 5 .
Value of Composite Reliability and Cronbach's Alpha.

Table 6 .
Path Coefficient and P-value.

Table 7 .
Indirect mediation results for path with 2 and 3 segments