The proposed study provides an insight into the consumer trust social networking sites, ultimately on user generated brand recommendations, and also investigating the role of ad-scepticism. The work contributes to a better understanding of trust development in social networking sites. Specifically, the study reveals that not all dimensions of trustworthiness are equal. The individual user characteristic varies according to the person. The major finding is that, high degrees of trust toward user generated brand recommendations can be generated on the basis of high trust toward the social networking sites and ad-scepticism. Consumer trust the user generated brand recommendations based on the individual’s trust in the particular social networking platform, and the level of ad-scepticism.

The independent variables used in this study are benevolence, integrity, ability, propensity to trust and individual user characteristic. The individual user characteristic used in the study is the consumer’s willingness to rely on user generated brand recommendations. The dependent variables are trust in social networking sites, ad - scepticism and trust in user generated brand recommendations. Trust in social networking sites is measured in this study as the consumers trust in their instagram friends. Benevolence is the kindness that a customer possesses while recommending a brand to another customer. Integrity is the measure of fairness that a customer possesses while recommending a brand to another customer. Ability is defined as the skill that a customer possesses while recommending a brand to another customer. Propensity to trust is described as the trust that a customer possesses while recommending a brand to another customer. Individual user characteristics are personality, demographics, and use behaviour of a customer who recommends a brand to another customer. Trust towards Instagram followers is the trust that a customer possesses while a brand is recommended by another customer who is a follower of him/her on Instagram. Ad-scepticism is the tendency toward disbelief of advertising claims that a brand is demanding. Trust in user generated brand recommendations is the trust towards the user generated brand recommendations on social networking sites. 

Integrity is found to be highly correlated with ability. The coefficient of correlation between trust in user generated brand recommendations and ability is only 0.445. The variables propensity to trust and trust in user generated brand recommendations show a moderate positive correlation between each other with a correlation coefficient of 0.572. The variable benevolence shows lowest correlation with propensity to trust. Among the variables under study, ability has recorded highest correlation with Trust in user generated brand recommendations with coefficient of correlation 0.645. The coefficient of correlation between trust in social networking sites and trust in user generated brand recommendations is calculated as 0.508. The variables ad-scepticism and trust in social networking sites show a moderate positive correlation between each other with a reported correlation coefficient of 0.623. The variables propensity to trust shows only low correlation with ad-scepticism. The variables trust in social networking sites and ad-scepticism are moderately correlated with each other. There exists a low positive correlation between ability and trust in social networking sites. A Goodness of Fit Index (GFI) value of 0.987 indicates that the model is appropriate. The GFI value increases and decreases according to the number of parameters. The Adjusted Goodness of Fit Index (AGFI) value of 0.88 is also appropriate. Here the sample size of 281 has increased the AGFI value. CMIN is the relative chi-square. It is the chi-square fit index divided by the degrees of freedom and represents the minimum value of the discrepancy. In the case of maximum likelihood estimation, CMIN contains the chi-square statistic. The chi-square statistic is an overall measure of how many of the implied moments and sample moments differ. The more the implied and sample moments differ, the bigger the chi-square statistic, and the stronger the evidence against the null hypothesis. The value should be between 1 and 5. Here the CMIN value is 3.83, which is an acceptable fit. The Root Mean Square Error of Approximation (RMSEA) tells us how well the model, with unknown but optimally chosen parameter estimates would fit the population’s covariance matrix. In a well-fitting model the lower limit is close to 0 while the upper limit should be less than 0.08. Here the value is 0.07 which is close to the upper limit