Global Journal of Management and Business Research, E: Marketing, Volume 22 Issue 3

completed online questionnaires. All construct measures were developed following the precise definitions found in the literature review. The variables utilized to assess contextual factors, brand community participation, brand awareness, brand association, and brand image were derived from past empirical research. Brand community participation has been measured utilizing six items adopted from the study by (Langaro et al., 2019). Besides, brand awareness, brand image, and brand association have been measured using five, three, and five items consecutively adopted from the study (Sasmita & Mohd Suki, 2015). The items are available in the appendix section. The perceptions of the respondents regarding the items have been evaluated on a 5 ‐ point Likert scale ranging from 1(strongly disagree) to 5 (strongly agree). The researcher used descriptive statistics with SPSS version 25. SmartPLS3 was used to evaluate the hypothesis and conceptual model using partial least squares (PLS) path modeling. We used SEM with PLS to test ideas and analyze data. According to Ringle et al. (2020), scholars frequently use PLS-SEM. However, PLS-SEM and CB-SEM are two SEM estimation techniques (Astrachan et al., 2014). PLS-SEM focuses on explaining variance in endogenous variables, whereas CB-SEM estimates the covariance matrix and ignores explaining variance (Hair et al., 2017). PLS-SEM has many advantages over CB-SEM. PLS-SEM develops a theory and makes a prediction, whereas CB- SEM evaluates an existing idea. Second, PLS-SEM can explore formative and reflective interactions, whereas CB-SEM only analyzes reflective ones (Hair et al., 2017). Richter et al. (2016) say PLS-SEM can handle intricate models with multiple indicators and constructs, is handy with small sample sizes, and give valuable latent variable scores for research. IV. A nalysis and F indings Particulars N % Particulars N % Gender Yearly frequency of buying brands Male 285 64.04 Occasionally 235 52.81 Female 157 35.28 Seldom 101 22.70 Prefer not to say 2 0.45 Frequently 78 17.53 Education When necessary 31 06.97 MPhil/PhD 17 3.82 Yearly expenditure on brands Postgraduate/Masters 199 44.72 More than 10000 110 24.72 Undergraduate/Honors 167 37.52 6001-10000 98 22.02 HSC 62 13.93 2001- 6000 213 47.87 Profession Less than 2000 18 4.04 Student 154 34.61 Other amounts 6 1.35 Private service 103 23.15 Favorite fashion brands Govt. service 78 17.53 Aarong 155 34.83 Business/Self-employed 67 15.60 Cats Eye 86 19.33 Unemployed 27 6.07 Ecstasy 43 9.67 Housewife 16 3.60 Gentle Park 35 7.87 © 2022 Global Journals Global Journal of Management and Business Research Volume XXII Issue III Version I Year 2022 ( ) E 75 The following are the demographic characteristics of the participants (n = 445): 64.04% (n = 285) of the responders were male, whereas 35.28% (n = 157) were female. Other information regarding the respondents is provided in Table 1. c) Measures d) Data Preparation and Analysis Techniques a) Respondents Profile Anjan’s, Sailor, Zara Fashion, Gentle Park, Cats Eye, Ecstasy, Kay Kraft, Dorjibari, Le Re Ve, etc. in the preceding year. We only took into account the brand that sells items for both men and women of all ages. Following a thorough literature assessment, a complete questionnaire with trustworthy survey items was designed. To collect the data, Google Forms was used to develop an online questionnaire. The questionnaire was written in the English language. It was voluntary to take part in this research. From June 21, 2022 until September 21, 2022 data was collected. The researchers shared the questionnaire online with at least 650 individuals as an online survey using email and social networking platforms. However, the total number of individuals reached by the researcher is unknown. After finishing the study, the researcher found 445 Impact of Social Media-Based Brand Community Participation on Brand Image in Bangladesh: Mediating Role of Brand Association and Brand Awareness