Factors affecting the Gender- based Consumer purchase intention towards Ethical Fashion (A case study on undergraduates of the University of Sri Jayewardenepura)

Vidyodaya Journal of Humanities and Social Sciences VJHSS (2023), Vol. 08 (02) Factors affecting the Gender- based Consumer purchase intention towards Ethical Fashion (A case study on undergraduates of the University of Sri Jayewardenepura) P. A. D. N. N. Wijeratne+ and D. P. K. Manel Department of Social Statistics, Faculty of Humanities and Social Sciences, Sri Lanka Article Info Article History: Received 28 Feb 2023 Accepted 30 June 2023 Issue Published Online 01 July 2023 Key Words: Consumer Purchase Intention Ethical Fashion Gender *Corresponding author E-mail address: nimfiyanouchelli@gmail.com https://orcid.org/00090005-6285-6300 Journal homepage: http://journals.sjp.ac.lk/in dex.php/vjhss http://doi.org/10.31357/fh ss/vjhss.v08i02.14 VJHSS (2023), Vol. 08 (02), pp. 193-210 ISSN 1391-1937/ISSN 2651-0367 (Online) Faculty of Humanities and Social Sciences 2023 ABSTRACT The study examines the factors affecting the gender-based consumer purchase intention towards ethical fashion in Sri Lanka. Primary data was gathered using self-administrated questionnaire through online platform and physically. Sample of the study was selected using stratified and simple random sampling techniques through undergraduates from three selected faculties of the University of Sri Jayewardenepura. Sample size of the study was 371 respondents including 228 female and 143 male undergraduates. Both descriptive analysis and Structural Equation Modelling (SEM) method were applied for the quantitative data. Findings of the study revealed that both the female and male consumers have an identical level of awareness regarding the ethical fashion market and related scopes. The respondents of the study have a positive impression on ethical fashion concept, and they believe that the concept itself is necessary for the betterment of the society and environment. Attitudes and beliefs regarding ethical fashion and subjective norms were identified as the major factors which affect the purchase intention towards ethical fashion among the consumers. Furthermore, it was found that gender doesn’t have any moderate effect in determining the consumers’ purchase intention towards ethical fashion in the young adult consumers in Sri Lanka. Findings of the study suggested that the fashion industry should implement new methods to motivate the purchasing intention of young customers. Also, such methods should promote the core values of ethical fashion concept as most of the consumers are not aware about those and for the expansion of the ethical fashion industry. Wijeratne & Manel, VJHSS (2023) Vol. 08 (02) pp. 193-210 humans as well. In fashion industry, there are over 40 million garment workers globally and an estimated 2% of them are only getting paid a living wage (Made Easy, 2020). The minimum pay a worker receives in the majority of the nations that produce cloth— Bangladesh, India, Sri Lanka, and China— represents between 20% and 50% of their living wage. 1. Introduction The topics of Sustainability and Ethics have become two of the majors focusing areas of the business world nowadays. Environmental damage caused by improper practices throughout the manufacturing process, Inhumane working conditions, degradation of social values have led the globe to focus on the sustainable and ethical business practices and consumption. Additionally, workers who operate in hazardous buildings without ventilation, risk breathing in poisonous materials or fiber dust. The most devastating accident happened in the garment industry was in 2013, when the garment building “Rana Plaza” in Dhaka, Bangladesh collapsed killing 1134 garment workers when the accident could have been avoided. (“Inhumane Working Conditions,” 2022). The clothing industry uses child labor extensively (160 million children are compelled to work). Ethical fashion companies try to shift this situation into a betterment. Sustainability combines environmental health, social justice, and economic vitality to create a prosperous and resilient community for present and future generations (UCLA, 2016). Ethical practices are necessary to achieve sustainability, including in the fashion industry, which is a dominant global business. Ethical Fashion is a new trend that promotes sustainability in the fashion industry. Ethical Fashion aims to minimize negative impacts on people, animals, and the environment throughout the entire garment manufacturing process (Made Easy, 2020). It considers aspects such as fair wages, sustainable production, and avoiding issues like child labor and modern slavery (Pollari, 2016). The fashion industry is particularly exposed to modern slavery risk (Keegan et al., 2021), and ethical practices are needed to address these issues. The fashion industry is one of the environmental polluters in the world, and is only second to the oil industry. Water pollution, immense water consumption, microfibers pollution, waste problem, chemical use in the industry, greenhouse gas emissions, soil and rainforest depredation are only a few of the hazardous environmental impacts caused by the fashion industry. The industry is responsible for the emission of about 2.1 million metric tons of greenhouse gases in 2018, which is about 4% of the total (Berg et al., 2020). As a response to this climate change due to the fashion industry, ethical fashion brands, companies are adopting circular business models. Especially Circular Fashion: this recognizes the value of used clothing as a material resource that can be used to recreate further value in the form of new goods. Simply it’s about recycling (Keegan et al., 2021). Ethical fashion brands aim to address various issues in the fashion industry such as fair wages, child labor, ecologically sustainable production, and animal welfare. The fashion industry is also responsible for the brutal death of millions of animals to obtain animal fur, skin, wool, and such for clothing production, which causes environmental pollution (YJ, 2018) (“Animals Used for Clothing,” 2022). Ethical fashion brands produce animal cruelty-free fashion products to address this issue. Now these issues have drawn an intense public interest in consumers, manufacturers, In the definition, it stated that the ethical fashion aims to reduce the harm towards 194 Wijeratne & Manel, VJHSS (2023) Vol. 08 (02) pp. 193-210 retailers, and designers as well (Shen et al., 2012). In that case, consumers have become aware of the ethical consumption particularly in developed countries (Karim, 2020) thereby a growing segment of consumers consciously buy ethical or sustainable products, such as organic, fair trade, animal friendly, locally produced, or eco-friendly products (Crane, 2001). in Sri Lanka are quite impressive, it is essential to identify whether the Sri Lankan fashion consumers are aware of the ethical or sustainable fashion. According to the industry insiders, Sri Lankan fashion consumer patterns are slowly getting aware about the climate changes caused by the fashion industry as well as about the sustainable consumption (Trust, 2022). Ultimately choosing ethical and sustainable products leads to a better consumption pattern which will be beneficial both socially and environmentally. For the choice or the preference of a consumer towards a particular product, especially when it comes to fashion products, gender plays a major role. Thus, this study will focus on the factors which affect the gender-based preference towards ethical fashion, with reference to the undergraduates of the University of Sri Jayewardenepura. When it comes to the Fashion industry in Sri Lankan context, it is an industry that has been existing for three decades. Moreover, the Sri Lankan apparel industry manufactures goods for world famous fashion brands. Ethical practices are also being practiced in the Sri Lanka Apparel and Fashion Industry while meeting the global demand. And Sri Lanka adopted the concept of sustainable and ethical fashion in the 2000’s. According to the Sri Lanka Export Development Board website, following measures have been taken by the Sri Lankan fashion industry to meet the ethical practices; Collective effort in reducing carbon footprint of suppliers and manufacturers. Additionally, Sri Lanka is positioned as a preferable destination by recognizing it as a trustworthy supplier of adequate wages for employees and by completely avoiding utilizing child labor and forced labor in its business. Sri Lanka has a social duty when purchasing clothing. The garment industry in Sri Lanka uses the most advanced technologies, including the first "Green Garment Factory" in the world, which reduces water and energy use by 70% and by half, respectively. (Export Development Board, 2022). Campaigns like “Garments without Guilt” became a national success as they promoted the need of ethical and positive work conditions for the garment workers as well as condemned the unethical practices such as child labor. Instead, they educated the fashion community about positive steps that could be taken to uplift ethical and sustainable practices (Garcia, 2021). Even though the ethical and sustainable practices of the fashion industry 1.1 Objectives of the Study The growth and survival of a product depends on the preference and the beliefs consumers hold on to that particular product. In that case, gender difference makes a bigger impact on the preference towards that product. This is same for Ethical fashion products as well. When the attitudes and beliefs are better, the growth and the survival of Ethical fashion products can be ensured. Therefore, to adapt into manufacturing Ethical fashion products more, level of the awareness and the preference of the consumers specifically based on the gender difference in purchasing Ethical products should be identified. Therefore, the study focuses on investigating how the young consumer crowd in Sri Lanka would purchase Ethical Fashion products having undergraduates of the University of Sri Jayewardenepura as the reference. The research specifically studies males’ and females’ preference towards Ethical Fashion. Thus, the objective of this study is to explore how consumers make purchase decisions on ethical fashion products gender-wise. 195 Wijeratne & Manel, VJHSS (2023) Vol. 08 (02) pp. 193-210 prioritize social status over ethical considerations. This suggests that social prestige and appearance have a greater impact on men's purchasing decisions (Achabou et al., 2020). On the other hand, women tend to be more concerned about ethical and sustainable consumption due to their strong environmental attitudes and concerns (Meyers-Levy & Maheswaran, 1991). According to gender socialization theory, societal norms shape individuals' behaviors and expectations, with women exhibiting more pro-environmental attitudes and behaviors, such as being expressive, compassionate, nurturing, altruistic, and caregiving (Costa Pinto et al., 2014). In China, men are more willing than women to spend on co-friendly clothing, and female consumers prioritize "no animal skin use" fashion. Chinese men also exhibit greater concern for environmental issues such as air quality, reduced water usage, and garment labor welfare (Rahman et al., 2020). 1.2 Literature Review Ethical fashion is currently a prominent trend in the fashion industry which has led the consumers globally to focus upon the ethical aspect and sustainability of the fashion industry (Keegan et al., 2021). Even though several research studies have been done on this topic, there is only a little academic research related to the Sri Lankan context of ethical fashion (Dissanayake et al., 2017). Tung describes in his study that, if the manufactures or the fashion companies can adopt ethical fashion practices such as green marketing strategies like promoting the fashion brand among the consumers by emphasizing eco-friendly concepts, exposing the eco-friendly and ethical mannerism and practices, the company is more likely to attract consumers towards the fashion brand and improve the company’s reputation (Tung et al., 2017). Because consumers’ conscious traits are more concerned about the ethical fashion attributes like preventing sweatshop labor, animal cruelty free environment protection (B. Shen et al., 2012). Even outlining a simple message related to ethical fashion attributes within the fashion brand, company could lead the fashion consumers to get attracted to a brand rather than complex message (Lähdesmäki, 2022). Exactly opposite to this scenario, Karim Farhana has concluded in her research that, fashion consumers in Bangladesh are not concerned about whether the clothes they purchase are ecofriendly, CSR concerned, or labor welfare concerned even if Bangladesh itself is one of the pioneer countries in manufacturing clothes. The reason she has described is that the lack of awareness of ethical fashion among consumers (Karim, 2020). Many studies have shown that it is essential to make consumers aware of and educate about the ethical fashion (Achabou et al., 2020). Existing literature also suggests that consumers' attitudes towards green products strongly influence their purchase intentions for ethical fashion (Sreen et al., 2018). Additionally, beliefs and attitudes towards animal cruelty-free fashion have a positive impact on the purchase intentions of undergraduates in Sri Lanka. However, when it comes to fair-trade fashion, despite holding beliefs and attitudes, there is little impact on their desire to make purchases. This could be due to the unfavorable opinions and attitudes towards the fair-trade component of clothing in Sri Lanka, as the culture often views the garment sector and its treatment of employees negatively (Thilakaratne, 2020). Numerous studies have shown that consumers' willingness to purchase ethical fashion is primarily determined by their attitudes towards ethically produced fashion products and their environmental concerns (Dimagi & Herath, 2017; Tjokrosoeharto & Paramita, 2021; Weiner, 2017). The existing literature on gender differences in ethical fashion reveals both consistent and contradictory findings. In Italy, men are more likely to engage in "show off" behaviors and 196 Wijeratne & Manel, VJHSS (2023) Vol. 08 (02) pp. 193-210 Several studies have explored the factors influencing consumers' green preferences and their purchasing decisions in relation to environmental and ethical concerns. Traditionally, age, income, gender, and geography were considered important sociodemographic variables in understanding these preferences. However, recent research suggests that while these factors do have some influence, they do not determine consumers' green preferences conclusively (Sreen et al., 2018). A study conducted by Karim in 2020 discovered that friends and family play a significant role in influencing consumers' decisions when it comes to fashion items in Bangladesh. The influence of friends and peers of the same age on consumer behavior is attributed to social norms. Furthermore, multiple studies support the idea that subjective norms, including social influences and personal beliefs, significantly influence individuals' purchase intention for ethical fashion (Karim, 2020; Nam et al., 2017; Okur & Saricam, 2019; Sreen et al., 2018; Weiner, 2017). Product attributes can be categorized into price, product quality. And these product attributes have been identified as other variables which determine the purchase intention towards fashion items (Forsman & Madsen, 2017; Karim, 2020). Self-expressive benefits involve customers consciously choosing items based on their personal morals (Birtchnell et al., 2006). Social responsibility among customers has been found to influence their purchasing decisions (Pookulangara et al., 2011). The intention to purchase ethically produced fashion products is significantly influenced by ethical self-identity. Customers who prioritize ethics in their buying decisions are more motivated to purchase fashion items that are ethically made. The increasing awareness of the environmental impact of fashion products is also influencing consumers' intention to buy ethically produced fashion (Koh & Noh, 2009; Weiner, 2017). 2. Materials and Methods The conceptual framework is based on the literature review. There are five independent variables that have been identified on the dependent variable: consumer purchase intention towards ethical fashion. Gender is identified as the moderate variable in this study (Figure 1). A case study was used to identify the factors affecting the gender-based consumer purchase of ethical fashion. The data was collected by a self-administrated questionnaire distributed both online and physically in the university. The sample of the case study consisted of male and female undergraduates of the University of Sri Jayewardenepura representing the young consumer crowd in Sri Lanka. University undergraduates are easier to reach, and therefore, the sample consisted of undergraduates from three selected faculties of the university: Faculty of Humanities and Social Sciences, Faculty of Management Studies and Commerce, and Faculty of Applied Sciences. These three faculties were particularly selected as most of the undergraduates of the University of Sri Jayewardenepura study in those faculties. Thus, the conclusions drawn from the selected sample could be considered as relatively precise as the sample represents the target population, undergraduates of the University of Sri Jayewardenepura. The sample was selected using the multi staged sampling method. The researcher performed the analysis in two stages. A univariate analysis using descriptive analysis and Principal Component Analysis (PCA), and Advanced statistical Analysis using Independent Sample t test and Structural Equation Modeling method. 197 Wijeratne & Manel, VJHSS (2023) Vol. 08 (02) pp. 193-210 Figure 1. Conceptual Framework Source: Compiled by the Researcher, Field Survey, 2022 61.5% and the rest of 38.54% are male undergraduates. It can be identified that the response rate of female undergraduates as 86.69% while male undergraduates’ response rate is higher by 33.64% than the expected response rate. 3. Results and Discussion This section presents the summary measurements of the variables explained through tables and graphs. The 371 of total respondents of the sample survey consists of majority of female undergraduates with a Table 1. Summary measures of Respondents’ Age Mean Variance Minimum Maximum Mode 22.927 1.943 20.000 26.000 24 Skewness Kurtosis -0.16 -0.92 Source: Field Survey, 2022 0.92). When considering the nature of living of the respondents (figure 2), it is evident that the sample consists of respondents who cover most of the residential settings of the country. Almost half of the respondents live in sub-urbanized areas and that’s 49.33% of the total responses. Most of the respondents are from the fourth year which is 41.24%, and 25.07% are from the first year, while According to the responses received as in table 1, mean age group is identified as 22– 23-year-old undergraduates. And undergraduates between 20 to 26 were included in the sample. Also, it can be identified that the distribution of age distributed with a slight negative skewness (-0.16) and negative kurtosis (198 Wijeratne & Manel, VJHSS (2023) Vol. 08 (02) pp. 193-210 respondents from the other two years have given a similar response percentage. 36.93% as it is the faculty with the highest number of undergraduates in the University of Sri Jayewardenepura. Respectively, Faculty of Humanities and Social Sciences and Faculty of Applied Sciences represent the response percentage 32.35% and 30.73%. When considering the family average Income depicted as in table 3 below, 32.88% of the respondents’ family average income is stated as above 90,000 Sri Lankan Rupees (LKR). The least number of respondents’ family average income is between 70,001 LKR and 90,000 LKR. Table 2. Frequency of Faculty-wise Respondents Faculty Percentage FHSS 32.35 FAS 30.73 FMSC 36.93 Source: Field Survey, 2022 Table 2. Frequency of Faculty-wise Respondents Table 3. Frequency of Respondents’ Monthly Family Income Faculty Percentage FHSS 32.35 FAS 30.73 FMSC 36.93 Source: Field Survey, 2022 Income Level Percentage Below 30,000 17.25 30,001-50,000 21.29 50,001-70,000 15.36 70,001-90,000 13.21 Above 90,000 32.88 Source: Field Survey, 2022 Table 2. Frequency of Faculty-wise Respondents Faculty Percentage FHSS 32.35 FAS 30.73 FMSC 36.93 Source: Field Survey, 2022 When it was investigated how often the respondents shop for clothes in six months of time, it was identified that most of the respondents ,70.6% shop for about 1-3 times. Most of the respondents, 68.46% have responded as they’re aware of Ethical Fashion. According to table 2, majority of the respondents are from the Faculty of Management Studies and Commerce with a 19% 32% Urban Semi-urban Rural 49% Figure 2. Composition of nature of Residence Source: Field Survey, 2022 199 Wijeratne & Manel, VJHSS (2023) Vol. 08 (02) pp. 193-210 Through Friends & Family 6% 5% 22% Through Internet Through Social Media Through Printed Media Through Radio 34% 33% Figure 3. Pie Chart for the type of source for the awareness (Source: Field Survey, 2022) Above figure 3, depicts how the respondents who are aware about Ethical Fashion are aware about it. It is evident that social media and the Internet have contributed more for the awareness among the respondents. distributes with a mean of 66.0277 which depicts the average level of purchase intention of respondents. Table 5. Summary Measurements of Purchase Intention Table 4. Component Score Coefficient Matrix Summary Measurement Mean Median Variance Std. Deviation Maximum Interquartile Range Skewness Absolute Skewness Kurtosis Absolute Kurtosis Source: Field Survey, 2022 Statement Component Score PUR_INT1 0.193 PUR_INT2 0.199 PUR_INT3 0.204 PUR_INT4 0.193 PUR_INT5 0.198 PUR_INT6 0.184 Source: Field Survey, 2022 The above component scores depicted in table 4 are extracted using PCA method. Based on these coefficients, factor scores are calculated for constructing the index. The summary measurements of the index are given by the table 5. Std. Error 66.0277 .91770 66.8967 312.445 17.67612 100.00 16.58 -.728 .127 -5.7323 1.412 .253 5.5810 Statistic It can be assumed that the index of purchase intention distributes in a normal distribution as it comes from a large sample size of 371, because the absolute skewness and kurtosis value falls between +7 and -7 (Kim, 2013). According to the summary measurements given in the above table, it can be identified that the purchase intention of the respondent 200 Wijeratne & Manel, VJHSS (2023) Vol. 08 (02) pp. 193-210 Following hypothesis can be built to conduct the t test. H0: Purchase Intention and Gender are independent. H1: Purchase Intention and Gender are not independent. Measurement Model A measurement model measures the composite variables or latent variables (Fan et al., 2016). Figure 6 illustrates the initial measurement model or model for Confirmatory Factor Analysis (CFA). There are 26 observed variables, and five latent variables presented in the diagram. Three observed variables were removed from awareness as the model validation of the measurement model was violated when those three variables were incorporated to the model. Fit describes a model's capacity to reproduce the data. A good-fitting model is one that reasonably matches the data and hence does not necessitate re-specification. Fit describes a model's capacity to replicate the data. Good-fitting models don't always need to be re-specified because they are reasonably consistent with the data. Before analyzing the structural model's causal paths, a well-fitting measurement model is also necessary. The above table 6 depicts the results of independent sample t test conducted to measure the effect of gender on purchase intention towards ethical fashion. According to the shown result, the null hypothesis is not rejected as the significance value exceeds the alpha value (0.648 >0.05). Therefore, it can be identified that there’s no significant effect of gender on the purchase intention towards ethical fashion. According to table 7, the KMO measure verified the sampling adequacy for the analysis and the KMO index was greater than 0.7; all KMO values for individual items were >0.85, which is above the desired level of 0.70. Barlett’s test of sphericity for each construct was statistically significant (p < 0.00)(Hoyle, 2012; Yan Piaw, 2023). Table 6. Results of independent sample t test for Purchase Intention vs Gender t df Sig. (2tailed) Mean Difference -.457 369 .648 -.463 313.559 .644 Source: Field Survey, 2022 -.86280 -.86280 Std. Error Difference 1.88757 1.86497 95% Confidence Interval of the Difference Lower Upper -4.57454 2.84894 -4.53224 2.80664 Table 7. Reliability and Validity Measures Construct Purchase Intention Consumer Awareness Attitudes & Beliefs Subjective Norms Self-expressive Benefits Cronbach's Alpha Kaiser-Meyer-Olkin Measure of Sampling Adequacy 0.925 0.862 0.897 0.913 0.917 0.909 0.887 0.889 Bartlett's Test of Sphericity ChiPSquare value 1573.521 0.000 1104.583 0.000 1048.169 0.000 1502.906 0.000 0.901 0.839 928.140 201 0.000 Wijeratne & Manel, VJHSS (2023) Vol. 08 (02) pp. 193-210 Sources: Sample Survey, 2022 Figure 4: Measurement Model (Source: Field Survey, 2022) Table 8. Results of Goodness of Fitness of the Measurement Model Goodness of fit Index Absolute fit indices Chi Square Index (CMIN/DF) Goodness of statistic (GFI) Adjusted Goodness of Fitness statistic (AGFI) 202 Observed value 2.29 0.876 0.849 Accepta ble value <5 0–1 0–1 Wijeratne & Manel, VJHSS (2023) Vol. 08 (02) pp. 193-210 Root Mean Square Error of Approximation (RMSEA) Tukey-Lewis Index (TLI) Comparative Fit Index (CFI) Incremental fit indices Relative Fit Index (RFI) Normed Fit Index (NFI) Parsimony Goodness of Fit Index (PGFI) PRATIO Parsimony fit indices Parsimonious Normed Fit Index (PNFI) PCFI Source: Sample Survey,2022 Once the overall model fit is considered, CMIN/DF recorded 2.29, making the measurement model acceptable as to minimize dependence on sample size, the suggested acceptable value for relative CMIN/DF, should be as high as 5 (Maat et al., 2015). As shown in Table 8, the RMSEA, which assesses the hypothesized model fit with a population covariance matrix, is 0.059 for the estimated model while GFI that represent the overall amount of the covariation among the observed variables that can be accounted for by the model is 0.876. The CFI and the value for the model are greater than 0.9 (CFI = 0.942), indicating a good overall fit of the measurement model. The NFI value of this study was 0.901 which is greater than 0.9. The TLI and RFI values for this model were 0.934 and 0.889 respectively, and it indicates a good incremental fit. 0.059 < 0.1 0.934 0.942 0.889 0.901 0.721 0.889 0.802 0.837 0–1 0–1 0–1 0–1 0–1 0–1 0–1 0–1 Moreover, Parsimony-Adjusted Measures are also close to one. As a result, every model fit index satisfies the criteria for a measurement model that fits well. Validation of the Measurement Model The validity of the measurement model can be assessed by testing its discriminant and convergent validity. Convergent validity allows for the measurement of a certain construct using variables that converge and share a high proportion of variance. In this study, three criteria—individual standardized factor loadings, Average Variance Extracted (AVE), and composite reliability—were used to assess convergent validity. Table 9. Results of the Convergent Validity Test Construct Purchase Intention (PURINT) Awareness (AWA) Subjective Norms (SUBNOR) Attitudes & Beliefs (ATTBEL) Self-Expressive Benefits (SELEXP) Source: Field Survey,2022 No. of items 6 Standardized factor loadings Min Max 0.746 0.877 Average Variance Exacted 0.674 Composite Reliability 0.925 5 6 0.635 0.790 0.812 0.859 0.527 0.641 0.847 0.914 5 0.723 0.867 0.637 0.897 4 0.764 0.865 0.699 0.903 203 Wijeratne & Manel, VJHSS (2023) Vol. 08 (02) pp. 193-210 For reflective indicators, the appropriate level of standardized factor loadings is equal to or greater than 0.70 (Hair et al., 2010). Table 9 indicates that standardized factor loadings are greater than or closer to 0.7 and they are significant at 5%. Also, it indicates that all the AVE values are greater than or closer to 0.6 and all the composite reliability measures are greater than 0.7. Hence, it can be concluded that the convergent validity requirements are satisfied in this scenario. To determine the discriminant validity of each construct, the square root of AVE of each construct was compared to the interconstruct correlation estimations of each construct. The estimate of the inter-construct correlations between each construct and all other constructs should be lower than the square root of AVE for all constructs. Table 10 compares the squared inter-construct correlations estimates with the AVE for all constructs. AVE for all constructs is presented in the diagonal entries bolded in table 10 which presents Fornell Larcker criterion evaluation of discriminant validity and the squared inter-construct correlations estimates among constructs are presented in the sub-diagonal entries. Also, table 12 indicates that the AVE for each construct was higher than the squared correlations between that construct and other constructs. Therefore, it can be concluded that the discriminant validity requirement is satisfied. Table 10. Comparison of Squared Inter-construct Correlations with AVE SELEXP AWA SUBNOR ATTBEL PURINT SELEXP 0.836 0.084 0.223 0.350 0.422 AWA 0.726 0.093 0.215 0.206 SUBNOR ATTBEL PURINT 0.801 0.506 0.400 0.798 0.745 0.821 Source: Sample Survey,2022 Figure 5. Main Structural Model (1st Structural Model) (Source: Compiled by the researcher using field survey data, 2022) 204 Wijeratne & Manel, VJHSS (2023) Vol. 08 (02) pp. 193-210 relationship between purchase intention and other independent variables considering both male and female respondents. Identify the Direct Effects of Awareness, Attitudes & Beliefs, Subjective Norms and Self-expressive Benefits on Purchase Intention Goodness of Fitness of the Main Structural Model In a theoretical model, the structural model depicts the causal and correlational relationships between latent variables. To identify the direct effect of consumer awareness, subjective norms, attitudes & beliefs, and self-expressive benefits on both male and female consumer purchase intention towards ethical fashion, the main structural model was developed. Below table depicts the summarized results of the goodness of fit of the main structural model. The overall goodness of fit in the main structural model can be verified as the CMIN/DF value is less than 5 and RMESA value is less than 0.1 and other goodness of fit indices are close to one. Therefore, the main structural model can be used to measure the direct and indirect effect of the identified independent variables (awareness, subjective norms, attitudes & beliefs, and selfexpressive benefits) on the purchase intention towards ethical fashion. This structural model is comprised of five major latent constructs with purchase intention as the dependent variable, and the others as independent variables. Figure 7 illustrates the structural model for direct Table 11. Results of the Goodness of Fit of the Main Structural Model Goodness of fit Index Absolute fit indices Incremental fit indices Parsimony fit indices CMIN/DF GFI AGFI RMSEA TLI CFI RFI NFI PGFI PRATIO PNFI PCFI Observed value 3.104 0.840 0.806 0.075 0.893 0.905 0.850 0.866 0.694 0.892 0.773 0.807 Acceptable value <5 0–1 0–1 < 0.1 0–1 0–1 0–1 0–1 0–1 0–1 0–1 0–1 Source: Field Survey, 2022 Table 12. Results of Regression Weights Path Path Coefficients AWA→PURINT SUBNOR→PURINT ATTBEL→PURINT SELEXP→ PURINT Source: Sample Survey, 2022 -0.10 -0.276 1.511 0.000 Standardized Path Coefficients -0.097 -0.315 1.186 0.000 205 Standard Error C.R Pvalue 0.046 0.048 0.134 0.038 -2.198 -5.787 11.277 0.006 0.350 0.001 0.001 0.946 Wijeratne & Manel, VJHSS (2023) Vol. 08 (02) pp. 193-210 Figure 6. Structural Model for Moderation for female respondents (Source: Compiled by the researcher using field survey data, 2022) benefits have no significant effect on purchase intention towards ethical fashion in both male and female consumers as the p values are greater than 0.05 (awareness = 0.350, self-expressive benefits = 0.946). Path coefficients of the Structural Model Table 12 depicts the results of the regression weights which are used to identify the direct effects of the constructs in both male and female respondents to measure the overall effect. According to the table, the subjective norms have a significant negative relationship with purchase intention (𝛽 = 0.276, P = 0.001). Therefore, it can be concluded that subjective norms negatively affect consumer purchase intention towards ethical fashion in both male and female consumers. Also, the table indicates that the attitudes & beliefs have a significant positive relationship with purchase intention (𝛽 = 1.511, P = 0.001). It implies that attitudes & beliefs have a positive effect on consumer purchase intention towards ethical fashion in both male and female consumers. Furthermore, it presents that the other two constructs, awareness and self-expressive Moderating Effect of Gender Figure 8 illustrates the structural models developed to identify the moderating effect of gender (female consumers) for identifying independent variables on purchase intention. The results of the goodness of fit for the structural models of female consumers indicates that CMIN/DF values are less than five and RMESA values are less than 0.1 and other goodness of fit indices are close to one in the results of goodness of fit for female consumers. Therefore, it can be considered that the structural models for female consumers are validate for further analysis. 206 Wijeratne & Manel, VJHSS (2023) Vol. 08 (02) pp. 193-210 Table 13. Results of the Goodness of Fit of the Structural Model for Moderation of female respondents Goodness of fit Index CMIN/DF GFI Absolute fit indices AGFI RMSEA TLI CFI Incremental fit indices RFI NFI PGFI PRATIO Parsimony fit indices PNFI PCFI Source: Sample Survey, 2022 Observed value 2.204 0.823 0.784 0.073 0.903 0.914 0.836 0.855 0.675 0.886 0.758 0.810 Acceptable value <5 0–1 0–1 < 0.1 0–1 0–1 0–1 0–1 0–1 0–1 0–1 0–1 Table 14. Results of Regression Weights of Moderation of female respondents Path Coefficients AWA→PURINT 0.036 SUBNOR→PURINT -0.153 ATTBEL→PURINT 1.166 SELEXP→ PURINT 0.082 Source: Field Survey, 2022 Path Standardized Path Coefficients 0.035 -0.168 1.039 0.097 Standard Error 0.043 0.044 0.109 0.037 C.R 0.839 -3.452 10.678 2.253 Pvalue 0.676 0.016 0.001 0.211 Figure 7. Structural Model for Moderation of male respondents (Source: Compiled by the researcher using field survey data, 2022) 207 Wijeratne & Manel, VJHSS (2023) Vol. 08 (02) pp. 193-210 Figure 9 illustrates the structural models developed to identify the moderating effect of gender (male consumers) for identifying independent variables on purchase intention. The results of the goodness of fit for the structural models of male consumers indicates that CMIN/DF values are less than five and RMESA values are less than 0.1 and other goodness of fit indices are close to one in the results of goodness of fit for male consumers. Therefore, it can be considered that the structural models for male consumers are validate for further analysis. considering the gender of the consumer as the moderator. According to the two tables, it can be concluded with 95% confidence level that the impact of subjective norms and attitudes & beliefs on consumer purchase intention is significant for both females and males. Therefore, it implies that the gender does not have a moderating effect of the identified independent variables on consumer purchase intention. The results of the moderating effect of the gender of the consumers for the purchase intention towards ethical fashion presents that, the gender of the respondents has no significant moderating effect on the purchase intention towards ethical fashion at 95% confidence level. Table 15 and16 show the impact of independent variables (awareness, subjective norms, attitudes & beliefs, and selfexpressive benefits) on purchase intention of both male and female consumers by Table 15. Results of the Goodness of Fit of the Structural Model for Moderation of male respondents Goodness of fit Index Absolute fit indices Incremental fit indices Parsimony fit indices CMIN/DF GFI AGFI RMSEA TLI CFI RFI NFI PGFI PRATIO PNFI PCFI Observed value 2.100 0.774 0.726 0.088 0.858 0.873 0.760 0.786 0.639 0.892 0.701 0.779 Source: Field Survey, 2022 Acceptable value <5 0–1 0–1 < 0.1 0–1 0–1 0–1 0–1 0–1 0–1 0–1 0–1 Table 16. Results of Regression Weights of Moderation of male respondents Path Coefficients AWA→PURINT -0.533 SUBNOR→PURINT -0.628 ATTBEL→PURINT 2.566 SELEXP→ PURINT -0.240 Source: Sample Survey, 2022 Path Standardized Path Coefficients -0.512 -0.689 1.799 -0.291 208 Standard Error 0.164 0.170 0.497 0.124 C.R -3.245 -3.688 5.157 -1.926 Pvalue 0.138 0.008 0.001 0.173 Wijeratne & Manel, VJHSS (2023) Vol. 08 (02) pp. 193-210 are not aware about what this concept means and how it can contribute to the society and the environment. It is necessary to promote the concept more among the consumers in Sri Lanka. Specially, how these products are made, the current policies and regulations followed in the fashion industry to produce ethical fashion products. As the consumers’ attitudes and beliefs towards ethical fashion were significant in leading them to purchase ethical fashion products, it is essential to leave a strong positive impression in the consumers about the ethical fashion concept for these products to be purchased more. That can be done through proper awareness and promotion of ethical fashion products and ethical fashion brands. According to the findings, as these consumers receive awareness about this concept mostly through social media and internet, ethical fashion producers and marketers can use social media platforms more to promote their brands and products. Another recommendation is that, since gender does not affect the purchase intention towards ethical fashion, this concept can be promoted between both male and female consumers equally. Consumers are also concerned about the affordability of these products. Therefore, when promoting ethical fashion products, it’s better to the make consumers aware about the affordability and that these products are worth buying. 4. Conclusion and Recommendation Findings of the study reveal several factors regarding purchase intention towards ethical fashion among the young adult consumer crowd in Sri Lanka taking undergraduates of University of Sri Jayewardenepura as the reference. Respondents’ attitudes and beliefs on ethical fashion are identified as a major factor which affects their purchase intention towards ethical fashion. The better the attitudes and beliefs respondents have on ethical fashion, the more they are willing to purchase ethical fashion products. Also, it was identified that subjective norms also affect respondents’ purchase intention towards ethical fashion. Respondents’ thoughts of being a socially and environmentally responsible consumer drives them to purchase ethical fashion products more. It was identified that selfexpressive benefits are not significant in driving the consumers to have purchase intentions towards ethical fashion. The main consideration of this study was to identify how gender plays its role in purchase intention of consumers towards ethical fashion. The findings of the bivariate analysis reveal that the gender doesn’t affect the purchase intention towards ethical fashion. Also, the advanced statistical analysis shows that gender is not a moderate factor which influences in having purchase intention towards ethical fashion. Even though literature shows that gender plays a major role in fashion choices among the consumers, what can be understood from this study is that, when it comes to ethical fashion, gender doesn’t affect the purchase intention. Both male and female consumers have similar thoughts on the concept. 5. References Fan, Y., Chen, J., Shirkey, G., John, R., Wu, S. R., Park, H., & Shao, C. (2016). Applications of structural equation modeling (SEM) in ecological studies: An updated review. Ecological Processes, 5(1), 19. https://doi.org/10.1186/s13717-0160063-3 Hair, J. F., Black, W. C., Babin, B. J., & Anderson, R. E. (Eds.). (2010). Multivariate data analysis: A global perspective (7. ed., global ed). Pearson. Based on the findings of the study, following recommendations regarding the ethical fashion industry in Sri Lanka can be made. Even though the majority of the consumers know the term ethical fashion concept, they 209 Wijeratne & Manel, VJHSS (2023) Vol. 08 (02) pp. 193-210 Hoyle, R. (2012). Handbook of Structural Equation Modeling. The Guilford Press, A Division of Guilford Publication, Inc. 72,Spring Street, New York. file:///C:/Users/HP/Downloads/Handbo okofStructuralEquationModeling.pdf Kim, H.-Y. (2013). Statistical notes for clinical researchers: Assessing normal distribution (2) using skewness and kurtosis. Restorative Dentistry & Endodontics, 38(1), 52–54. Maat, S. M., Adnan, M., Abdullah, M. F. N. L., Ahmad, C. N. C., & Puteh, M. (2015). Confirmatory Factor Analysis of Learning Environment Instrument among High Performance School Students. Creative Education, 6, 640–646. Made Easy, E. (2020). Every question you’ve ever had about ethical fashion, answered. [Organizational]. Ethical Made Easy. https://ethicalmadeeasy.com/what-isethical-fashion/ UCLA. (2016). UCLA Sustainability Charter. Sustainable Committee, UCLA. https://www.sustain.ucla.edu/wpcontent/uploads/UCLA-SustainabilityCharter.pdf Yan Piaw, C. (2023). A Step-by-step Guide, PLSSEM Data Analysis Using SMARTPLS4. Research tree Education. 210