The rapid diffusion of smartphones and digital platforms has transformed how individuals communicate, access information, and make consumption decisions. In this environment, social media has evolved from a communication tool into a major marketing channel through which firms can reach consumers quickly, interactively, and at a relatively low cost. In Indonesia, this transformation is especially significant because digital adoption is already at a large scale. The Indonesian Internet Service Providers Association reported that the country had 221.6 million Internet users in 2024, equivalent to an Internet penetration rate of 79.5 percent. In addition, DataReportal (2025a) estimated that Indonesia had 143 million active social media user identities in January 2025, demonstrating the centrality of social media in their daily lives.
Among social media platforms, Instagram occupies a particularly relevant position in advertising research. DataReportal (2025b) reported that Instagram had 103 million users in Indonesia in early 2025, and its advertising reach was equivalent to 48.7 percent of the country’s internet user base. Beyond mere usage volume, social media plays a substantial role in product and brand discovery. DataReportal (2025b) found that 66.7 percent of adult social media users in Indonesia use social platforms to learn about brands. These figures indicate that Instagram is not only a highly visible platform but also a strategically important environment in which advertising messages can shape consumer perceptions and behavioral intentions.
The growing importance of social media in marketing has been widely acknowledged. Prior studies have shown that social media advertising can influence purchase intention by increasing consumers’ exposure to product information, improving perceived relevance, stimulating entertainment value, and encouraging interaction with brands. Alalwan (2018), for example, demonstrated that core features of social media advertising are important predictors of customer purchase intention. Similarly, Hanaysha (2022) found that interactivity, informativeness, entertainment, and perceived relevance contribute positively to brand engagement, which in turn strengthens purchase intention. Hussain et al. (2022) further argued that the value created through social media advertising, particularly through entertainment, aesthetic appeal, interactivity, and trendiness, can enhance consumer engagement and encourage favorable behavioral outcomes, including repurchase intentions and electronic word-of-mouth.
In the Indonesian context, the relationship between social media marketing and purchase intention has also received empirical support. Vidyanata (2022) found that social media marketing positively affects purchase intention through a value-based adoption framework, indicating that consumers’ perceptions of value are central to understanding whether exposure to social media content translates into buying interest. Similarly, Chrisniyanti and Fah (2022) reported that social media marketing activities had a positive and significant effect on the purchase intention of Indonesian young adults, with additional mediation through subjective norms, perceived behavioral control, brand awareness, and social brand engagement. These findings suggest that young Indonesian consumers are not passive recipients of social media content; rather, their purchase intentions are shaped by how advertising messages generate value, trust, engagement, and social influence.
This issue is particularly relevant for university students, who represent a digitally connected consumer segment and are routinely exposed to online promotions on mobile devices. As active social media users, students are likely to encounter sponsored posts, promotional videos, endorsements, and other forms of paid content embedded in their daily online activities. Therefore, examining students’ purchase intentions in response to social media advertising provides a useful way to assess the effectiveness of digital promotional strategies in a young and highly networked population. Simultaneously, the effect of advertising may vary across platforms, audiences, and local contexts, making context-specific research necessary rather than assuming that findings from other consumer groups apply universally.
Against this background, the present study investigates whether social media advertising significantly influences the purchase intention of students at the Bina Nusantara University. This study is specifically limited to students in DKI Jakarta and focuses on Instagram as the platform being examined. By focusing on a student population in an urban Indonesian setting, this study seeks to contribute empirical evidence to the growing literature on digital advertising effectiveness and clarify the extent to which social media advertisements shape consumer purchase intentions among university students.

Advertising is generally understood as paid, mediated communication from an identifiable source designed to persuade audiences to take present or future actions (Richards & Curran, 2002). In marketing scholarship, advertising is not limited to informing consumers about product availability; it also performs persuasive, reminder, and reinforcement functions that shape perceptions and behavioral responses over time. This broader understanding is important because the current digital environment has changed the form of advertising without altering its central purpose: influencing consumer cognition, attitudes, and actions. In other words, although the medium has evolved from conventional mass media to interactive platforms, advertising remains a strategic communication tool in the promotion mix.
The development of social media has substantially expanded the context in which advertisements operate. Kaplan and Haenlein (2010) define social media as a group of Internet-based applications built on the ideological and technological foundations of Web 2.0 that allow the creation and exchange of user-generated content. From a marketing perspective, social media is significant because it enables two-way interaction, rapid information dissemination, and continuous engagement between brands and consumers. Mangold and Faulds (2009) argue that social media should be viewed as a hybrid element of the promotion mix because it allows firms to communicate with consumers while simultaneously enabling consumers to communicate with each other. This dual structure distinguishes social media from traditional advertising channels and helps explain why it has become central to contemporary promotional strategies.
Within this broader environment, social media advertising refers to promotional content distributed through social networking platforms to introduce, display, and persuade users to consider or purchase products and services. What makes social media advertising distinctive is not simply its digital format but its embedding within users’ everyday social activities. Rather than appearing only in isolated commercial spaces, social media advertisements are integrated into timelines, stories, reels, and feeds, where users also interact with peers, influencers, and brands. This creates a setting in which advertising can be more personalized, visually attractive, interactive, and behaviorally targeted than conventional forms of promotion are. Consequently, social media advertising is often more immediate and engaging, especially for audiences who spend substantial time on platform-based media.
Prior research has identified several features that explain why social media advertising influences consumer responses. Alalwan (2018) found that entertainment, informativeness, irritation, credibility, and perceived relevance are important predictors of purchase intention in social media settings. Similarly, Kim and Ko (2012) showed that social media marketing activities enhance customer equity by strengthening the value-related and relational dimensions of brand evaluation. Hanaysha (2022) further demonstrated that social media advertising features such as informativeness, entertainment, interactivity, and perceived relevance positively affect brand engagement, which in turn contributes to purchase intention. These studies indicate that the effectiveness of social media advertising depends not only on exposure but also on how consumers evaluate the advertising experience itself. Advertisements that are useful, engaging, and relevant are more likely to generate favorable attitudes and stronger behavioral intentions toward the brand.
Purchase intention was the dependent construct examined in this study. In consumer behavior research, purchase intention is commonly defined as an individual’s conscious plan or willingness to buy a product or brand in the future (Spears & Singh, 2004). This concept is particularly important because it functions as a proximal indicator of likely buying behavior, even when actual purchases are not directly observed. In the context of advertising theory, purchase intention is often linked to hierarchical response models, such as attention, interest, desire, and action, which suggest that persuasive communication influences consumers progressively rather than instantaneously. From this perspective, advertising is expected to first attract attention, then build interest, generate desire, and ultimately lead to action. The stronger the positive response to the message, the more likely the purchase intention.
Therefore, the relationship between social media advertising and purchase intention is well-grounded in theory and empirical evidence. Social media platforms provide high-frequency exposure, visual persuasion, and interactive brand contact, all of which can strengthen consumers’ evaluative responses to products. Hussain et al. (2022) argue that social media advertising value is shaped by entertainment, aesthetic appeal, interactivity, and trendiness, and that these features can foster consumer-brand engagement and value co-creation, which subsequently supports purchase intention. In the Indonesian context, Moslehpour et al. (2022) found that social media marketing activities significantly affect purchase intention both directly and indirectly through trust and brand image. Their findings are especially relevant to the present study because they show that in Indonesia, digital promotional activity can influence consumers not only by informing them but also by building symbolic and relational value around a brand.
These insights are directly relevant to the present study, which examines whether social media advertising influences the purchase intentions of Bina Nusantara University students. The study’s emphasis on Instagram is theoretically appropriate because image-based and interactive platforms intensify the persuasive elements identified in prior research, especially entertainment, informativeness, and relevance. For university students, who are routinely exposed to platform-based promotional content in their everyday smartphone use, social media advertising may function as a meaningful determinant of purchase intention rather than mere background communication. Accordingly, this study positions social media advertising as an independent variable and purchase intention as a dependent variable. Based on the literature reviewed above, the proposed hypothesis is that social media advertising has a significant positive effect on the purchase intention of students at Bina Nusantara University.

This study employed a quantitative, cross-sectional design to examine the effect of social media advertising on the purchase intentions of students at Bina Nusantara University in Jakarta. A cross-sectional design is appropriate when researchers seek to measure exposure and outcome variables at a single point in time and evaluate the association between them in a defined population (Setia, 2016). In the present study, the independent variable was social media advertising, and the dependent variable was purchase intention. This design is consistent with the study’s objective of determining whether social media advertisements, particularly those encountered on Instagram, significantly influence the purchasing intentions of university students. This section also follows the submitted draft, which positioned Bina Nusantara University students as the target population and identified simple linear regression as the principal analytical technique.
The study population consisted of students enrolled at the Bina Nusantara University. This population was selected because university students represent a digitally connected consumer group with high exposure to smartphone-based communication, social networking platforms and online promotional content. As established in the Introduction and Literature Review, students are an appropriate unit of analysis for research on social media advertising because they are active users of Instagram and similar platforms and are therefore likely to encounter advertising content embedded in their everyday online activities. In line with the submitted draft, this study was geographically limited to the DKI Jakarta setting and conceptually focused on students who were exposed to social media advertisements and demonstrated purchase-related interest.
The sample was drawn using quota sampling, a non-probability sampling technique in which respondents are selected until the predefined categories or target proportions are satisfied. Quota sampling is frequently used in survey-based research when researchers aim to secure participation from relevant respondent groups under the practical constraints of time, access, and cost (Futri et al., 2022). Quota sampling was suitable for this study because the research specifically targeted students who used social media and were potentially influenced by advertising content. The procedure described in the submitted draft indicates that respondents were selected based on relevance to the study criteria rather than through random sampling. This approach supports feasibility in online fieldwork, although it also means that findings should be interpreted as representative of the targeted respondent segment rather than all university students in a strictly probabilistic sense (Andrade, 2020; Eysenbach, 2004).
Data were collected using a self-administered online questionnaire distributed through Google Forms, as described in the submitted draft. Online questionnaires are widely used because they are relatively efficient, low-cost, and capable of quickly reaching digitally active respondents (Ebert et al., 2018; Wright, 2005). For a study involving university students, this mode of administration is methodologically appropriate because the respondents are generally familiar with digital interfaces and can complete the questionnaire at their convenience. At the same time, online surveys must be reported carefully because they may be affected by self-selection bias, unknown response denominators, and incomplete information about non-respondents (Andrade, 2020; Eysenbach, 2004). Therefore, the administration of the questionnaire should be described transparently in terms of recruitment channels, voluntary participation, response screening, and treatment of incomplete submissions.
The questionnaire was structured to measure the two principal constructs of this study: social media advertising and purchase intention. In line with the Literature Review, the social media advertising construct refers to respondents’ perceptions of promotional content encountered on social media platforms, especially Instagram, including its informativeness, attractiveness, persuasiveness, and relevance. Prior studies have shown that such features are central to how users evaluate social media advertisements and whether these advertisements influence subsequent buying intentions (Alalwan, 2018; Hanaysha, 2022). The dependent variable, purchase intention, refers to the respondent’s expressed willingness or likelihood of buying a product after exposure to promotional content. This conceptualization is consistent with established work that treats purchase intention as a proximal indicator of prospective consumer behavior (Spears & Singh, 2004). Because both variables involve perceptions and attitudes, they were appropriately measured using Likert-type questionnaire items, which are commonly used for latent constructs in survey research (Jebb et al., 2021).
Following this logic, respondents were asked to indicate their level of agreement with a series of statements related to social media advertising exposure and purchase intentions. Individual item scores were then combined into composite scores for each variable, allowing the constructs to be treated analytically as scale-based measurements. In a study such as this, the use of a structured rating format enhances comparability across respondents and supports subsequent statistical analyses. Nevertheless, good questionnaire practice also requires that item wording be clear, non-leading, and directly related to the study objectives, and that the instrument be reviewed for clarity before broader distribution (Eysenbach, 2004; Jebb et al., 2021).
To test the proposed relationship between variables, this study used simple linear regression. Regression analysis is an appropriate technique when the purpose is to estimate the magnitude and direction of the relationship between one independent variable and one dependent variable. In this study, the regression model is expressed as

Y = α + βX + ε

where Y represents the purchase intention, X represents social media advertising, α is the intercept, β is the regression coefficient, and ε is the error term. The coefficient β indicates whether changes in perceived social media advertising are associated with changes in purchase intention. A statistically significant positive coefficient supports the study hypothesis that social media advertising positively affects the purchase intention of Bina Nusantara University students. This analytical approach is fully consistent with the submitted draft, which identified a simple linear regression as the main inferential procedure.
Before interpreting the regression results, several standard assumptions should be assessed, including linearity, independence of errors, normality of residuals, homoscedasticity, and absence of extreme outliers. These checks are important because violations may distort the coefficient estimates or weaken inferential validity (Roustaei, 2024). Descriptive statistics should also be reported to summarize respondent characteristics and the central tendency of the measured variables before estimating the regression model. Given the use of quota sampling and an online questionnaire, the results should be interpreted cautiously, with appropriate recognition that the design is strong for examining associations within the targeted student group but limited in terms of broader population generalization (Andrade, 2020; Futri et al., 2022).
The method adopted in this study is appropriate for addressing the research problem. A cross-sectional survey enables the researcher to gather data efficiently from a digitally active student population, the online questionnaire format matches respondents’ communication habits, quota sampling provides a feasible means of reaching relevant participants, and simple linear regression offers a clear statistical framework for testing the hypothesized influence of social media advertising on purchase intention. Thus, the methodological design is logically aligned with the study’s introduction, theoretical grounding, and hypothesis structure.

The research method utilized IBM Statistics or SPSS output calculations at a 5% significance level.  The methods used were Validity and reliability tests were performed on the two variables, namely, Social Media Advertising and Purchase Interest. Classical assumption tests were also conducted, including a residual normality test, a heteroscedasticity test, and a simple regression test. Regression analysis is a statistical method used to determine the pattern of relationships between independent and dependent variables (see Table 1).

Table 1. Data

At the most basic descriptive level, the response pattern shows a strong clustering around the upper-middle and upper categories of the scale. When the uploaded raw item matrix was read directly, most observations on both variables fell at 3 or 4, with relatively few responses at 1 or 2. This pattern indicates that respondents generally expressed moderately favorable to favorable perceptions of social media advertising and, at the same time, moderately positive purchase intention.
Based on the uploaded response table, the item means for the social media advertising variable were approximately 3.36 (X1), 3.41 (X2), 3.33 (X3), 3.40 (X4), and 3.34 (X5), while the purchase intention items averaged approximately 3.40 (Y1), 3.23 (Y2), 3.24 (Y3), and 3.37 (Y4). At the composite level, the average observed score for social media advertising was approximately 3.37, whereas the average observed score for purchase intention was approximately 3.31. These descriptive values suggest that the respondents did not evaluate social media advertising negatively; rather, they tended to report a generally favorable assessment of advertising exposure and a similarly positive tendency toward purchase-related interests.
The descriptive pattern is analytically important because it provides the first empirical indication that the study context is one in which social media advertising is already salient to respondents. In practical terms, the student sample appears accustomed to online promotional content and does not respond with blanket rejection. This is important because the study’s theoretical logic, developed in the Literature Review, assumes that social media advertising can influence purchase intention when the audience is attentive to and meaningfully engaged with advertising content. If the descriptive data had shown uniformly low evaluations of the advertising items, the subsequent regression effect would have been less likely. Instead, the general concentration of responses in the 3–4 range indicates that respondents perceived social media advertising as sufficiently relevant to affect their downstream attitudes. This descriptive tendency is broadly consistent with prior work showing that social media advertising can shape consumer reactions when users perceive the advertising environment as relevant, engaging, informative or entertaining.
Before testing the hypothesis, the uploaded SPSS output assessed the assumptions underlying the simple linear regression. This is methodologically necessary because linear regression relies on several foundational assumptions, including normality of residuals and homogeneity of error variance. If these assumptions are violated, coefficient estimates may still be produced, but the inferential interpretation of the model becomes less secure. Kim (2019) explains that residual analysis is central to validating a linear regression model, particularly because assumptions concerning normality and equality of variance influence the defensibility of the estimated relationship. Likewise, Baždarić et al. (2021) stress that proper reporting of regression results requires that the assumptions be checked rather than merely presumed. The uploaded file follows this logic by presenting normality and heteroscedasticity diagnostics before the main regression test.
The first diagnostic reported in the file was the residual normality test. Because the dataset contains more than 50 observations, the uploaded analysis appropriately relied on the Kolmogorov–Smirnov test rather than the Shapiro–Wilk test, as its stated decision rule for large samples. The reported significance value was 0.052, which exceeded the threshold of 0.05. Based on the decision rule included in the file, the null hypothesis of a normal residual distribution was accepted. Substantively, this means that the regression residuals were sufficiently close to a normal distribution to support the continued use of parametric linear regression analysis. Although a significance value of 0.052 lies close to the conventional cut-off, it remains on the acceptable side of the threshold used in this study. Thus, the normality test did not provide evidence of serious violations. This implies that the model’s inferential statistics may be interpreted with greater confidence, since the residual structure does not undermine the general suitability of the regression framework.
The second assumption test is the heteroscedasticity test, which assesses whether the variance of the residuals remains relatively constant across the levels of the independent variable. Unequal error variance can weaken the trustworthiness of standard errors and significance tests; therefore, it is a routine and necessary diagnostic step in regression analysis. In the uploaded output, the significance value for the Social Media Advertising variable is 0.130, which is greater than 0.05. Based on the decision rule stated in the file, the null hypothesis of no heteroscedasticity was accepted for all models. This indicates that the model does not exhibit a serious heteroscedasticity problem and that the spread of prediction errors is acceptably stable across the levels of the predictor. In practical terms, the absence of heteroscedasticity strengthens the analytical credibility of the regression results because it suggests that the observed relationship between social media advertising and purchase intention is not distorted by a systematic change in error variance.
After the classical assumptions were checked, the uploaded analysis evaluated the construct validity of the two study variables. In questionnaire-based research, item validity is essential because the regression model can only be as meaningful as the instrument used to measure constructs. If individual items fail to correlate adequately with the intended variable, any subsequent model estimated from those scores is conceptually weakened. In the uploaded file, validity was tested by comparing each item’s calculated correlation coefficient with an r-table threshold of 0.17. For the Social Media Advertising variable, all five items exceeded the threshold: X1 = 0.577, X2 = 0.731, X3 = 0.491, X4 = 0.648, and X5 = 0.692. Because each value was greater than the benchmark used in the study, all five items were retained and interpreted as valid indicators of the constructs. These coefficients suggest that each item contributed meaningfully to the broader concept of social media advertising, rather than functioning as unrelated or weakly aligned questionnaire content.
The validity evidence for the Purchase Intention variable was somewhat more uneven but still acceptable according to the study’s stated decision rule. The reported item correlations were Y1 = 0.377, Y2 = 0.501, Y3 = 0.565, and Y4 = 0.316, all of which remained above the r-table value of 0.17. Based on this, all four purchase intention items were classified as valid in the uploaded analysis. It is worth noting, however, that the validity values for Y1 and especially Y4 are more modest than those of the stronger social media advertising items. This does not invalidate the scale, but it does indicate that the dependent variable is measured with less uniform strength across items than the independent variables. However, because none of the reported values fell below the study’s acceptance threshold, the scale remained adequate for hypothesis testing within the boundaries of the present design. Thus, the uploaded findings support the conclusion that the questionnaire items were sufficiently aligned with their intended constructs to proceed with reliability analysis and regression modeling.
Following the validity tests, the analysis assessed internal consistency reliability using Cronbach’s alpha, which remains one of the most widely used indicators of scale reliability in survey studies. Tavakol and Dennick (2011) explained that Cronbach’s alpha is useful for evaluating the extent to which items within a scale measure the same underlying construct, even though alpha must be interpreted in light of the number of items and conceptual coherence. In the uploaded file, the Social Media Advertising scale produced a Cronbach’s alpha of 0.829, clearly exceeding the study’s acceptance threshold of 0.60. This is a strong result and indicates a high degree of internal consistency among the five advertisement items. In other words, respondents’ answers to the X items moved together in a sufficiently stable pattern to justify combining them into a single predictor variable. From a measurement standpoint, this is important because the explanatory power of the regression model depends heavily on whether the independent variable is measured consistently.
The Purchase Intention scale produced a Cronbach’s alpha of 0.656, which also exceeds the threshold of 0.60 used in the uploaded analysis and therefore qualifies as reliable by the study’s stated criteria. Compared with the advertising scale, the internal consistency of the dependent variable was lower, but it remained acceptable for basic explanatory research using a short four-item instrument. This result is consistent with the earlier validity pattern, in which the purchase intention items were valid but not as uniformly strong as the advertising items. Methodologically, this distinction matters because weaker reliability in the dependent variable can reduce the precision with which the model captures the outcome. However, the reliability remained above the accepted threshold and therefore did not prevent formal hypothesis testing. The reliability findings indicate that the instrument was psychometrically workable: the independent variable was measured strongly, and the dependent variable was measured adequately. This combination is sufficient for interpreting the subsequent regression model with reasonable confidence.
Once the instrument and assumptions were judged acceptable, the study proceeded to the central inferential test: simple linear regression analysis. The uploaded file reports a coefficient of determination (R²) of 0.357, meaning that 35.7% of the variance in the purchase intention of Bina Nusantara University students is explained by the Social Media Advertising variable. Conversely, the remaining 64.3% of the variance was attributed to other factors outside the present model. This is a significant result. This does not imply that social media advertising is the sole determinant of purchase intention; rather, it suggests that the variable has a substantial explanatory contribution while leaving room for other influences, such as product price, peer recommendations, brand image, trust, need recognition, lifestyle, prior product experience, and broader platform engagement. In behavioral and marketing research, a single-variable model that explains more than one-third of the outcome variance is not considered trivial. This indicates that social media advertising has real explanatory force, even if it does not exhaust the complexity of consumer decision-making among university students.
The hypothesis test further supports this conclusion. The uploaded SPSS output reports a significance value of 0.000 and a calculated t-value of 7.374, compared with a critical t-table value of 1.66. According to the study’s decision rule, both criteria lead to the rejection of the null hypothesis: the probability value is below the alpha threshold of 0.05, and the calculated t-value is much larger than the critical value. Accordingly, the study concludes that Social Media Advertising has a statistically significant effect on the Purchase Intention of Bina Nusantara University Students. This is the central empirical result of this study. It directly answers the research question raised in the Introduction and supports the hypothesis formulated in the Literature Review section. In substantive terms, this finding indicates that the more positively students evaluate social media advertising, the stronger their purchase intention tends to be. The effect is not merely descriptive or incidental; it is statistically strong enough to reject the assumption that no meaningful relationship exists between them. The practical direction of this relationship is clarified by the following reported regression equation:

Y = 1.628 + 0.497X

This equation shows a positive slope coefficient (b = 0.497), meaning that for every one-unit increase in Social Media Advertising, the predicted Purchase Intention score increases by 0.497 units. The intercept of 1.628 indicates the predicted baseline level of purchase intention when the advertising variable was set to zero in the regression model. Although a zero score may fall outside the observed response range of the questionnaire, the intercept remains a standard mathematical component of the linear model and anchors the prediction line. The substantively important element is the slope, which confirms that the relationship between social media advertising and purchase intention is positive, not negative. Thus, improved or more favorably perceived advertising is associated with higher students’ purchase intentions. This directional result aligns precisely with the conceptual expectations developed in the preceding sections of this study.
Interpreted alongside the Introduction, these results suggest that social media advertising has become more than a background feature of the digital environment for the student population. The Introduction section argues that smartphones and social media are deeply embedded in everyday life, while Instagram, in particular, functions as a prominent space for product discovery, information exposure, and online consumption. The present findings provide empirical support for this proposition. The observed positive regression coefficient and statistically significant test results indicate that social media advertisements are not merely seen by students; they are associated with stronger purchase intentions. In that sense, the Results section empirically substantiates the manuscript’s earlier claim that digital promotional activity can influence consumers in a university setting, particularly where social media use is already normalized and routine behavior.
The findings are also highly consistent with the theoretical framework established in the Literature Review. This section emphasized that advertising functions by attracting attention, building interest, stimulating desire, and encouraging action, while social media advertising enhances this process through interactivity, informativeness, entertainment, and relevance. The present results support this logic at the scale level. Although the uploaded dataset does not provide the verbatim wording of items X1–X5 and Y1–Y4, the validated and reliable measurement structure indicates that the survey succeeded in capturing coherent perceptions of social media advertising and coherent expressions of the purchase intention. The significant positive association between the two constructs supports the general AIDA-based expectation that favorable advertising exposure can move consumers further along the path from awareness to purchase-related intention. In other words, the findings are not isolated statistics; they fit the advertising mechanism anticipated by the conceptual framework.
These results align with those of prior empirical studies cited in this manuscript. Alalwan (2018) found that key features of social media advertising predict purchase intention, while Hanaysha (2022) reported that social media advertising characteristics contribute to brand engagement and, through that process, to purchase intention. Hussain et al. (2022) similarly showed that social media advertising drives consumer co-creation and purchase-related outcomes. Moslehpour et al. (2022) demonstrated that social media marketing activities significantly influence purchase intention in the Indonesian context. The present study does not replicate those models in all their complexity because it uses a simpler one-predictor regression design. Nevertheless, the findings are consistent across studies: more effective or positively perceived social media advertising is associated with stronger purchase-related responses. Therefore, the current results reinforce, rather than contradict, the broader empirical literature. They also add value by confirming this relationship, specifically among Bina Nusantara University students.
Simultaneously, the results invite a more careful interpretation than a simple statement of significance. The R² value of 0.357 means that social media advertising is important, but it is not sufficient on its own to fully explain students’ purchase intention. A large portion of the variance remains outside the model, which is plausible in consumer research. Students’ purchase intentions are likely shaped not only by advertising stimuli but also by financial considerations, product type, perceived need, peer influence, trust in the brand or seller, platform credibility, celebrity or influencer effects, and previous purchase experiences. Therefore, the contribution of the present model should be framed as meaningful but partial in nature. This is one of the strengths of the findings rather than a weakness: the results identify social media advertising as a statistically significant factor without exaggerating it as the sole driver of student buying intention. Such an interpretation is more analytically credible and better aligned with the multidimensional nature of consumer behavior, as described in the literature.
Another noteworthy aspect of the findings is the relative strength of the measurement model for independent variables. The Social Media Advertising scale not only demonstrated strong reliability (α = 0.829) but also showed consistently solid item validity coefficients, particularly for X2, X4, and X5. Although the specific content of these items is not included in the uploaded output, the pattern implies that respondents answered the advertising indicators with relatively stable internal logic. In contrast, the Purchase Intention scale, although acceptable, showed more moderate psychometric values. This suggests that students’ evaluations of advertising were more internally coherent than their self-reported purchase intentions. This pattern is understandable. Perceptions of advertising may be easier to report consistently than future-oriented intentions, which are often more situational and contingent on the context. Nevertheless, the reliability of the outcome measure remained acceptable; therefore, the regression results remained interpretable. This nuance strengthens the credibility of the analysis because it acknowledges the instrument’s actual profile rather than presenting both scales as equally strong when the uploaded output indicates otherwise.
The findings also have practical implications for social media-based promotional strategies. If stronger perceptions of social media advertising are associated with higher purchase intention, advertisers targeting university students should pay close attention to the quality of ad execution rather than assuming that exposure alone is sufficient. Prior literature suggests that social media advertisements become more effective when perceived as relevant, informative, entertaining, interactive, and credible. The present findings do not isolate these dimensions individually, but they support the broader managerial implication that improving the quality of the social media advertising experience can increase consumer receptivity. For a student audience that is heavily immersed in smartphone-based media use, poorly designed, unconvincing, or irrelevant advertising is unlikely to produce the same positive movement in purchase intention as advertising that is well aligned with users’ interests and platform expectations. Thus, the empirical results point toward a strategic lesson as well as a statistical one.
Although this section demonstrates a statistically significant and theoretically coherent relationship, the findings should be interpreted within the boundaries of the study design. The uploaded output reflects a cross-sectional, self-report, online questionnaire administered through Google Forms and analyzed using a quota-based sample of Bina Nusantara University students. Such a design is suitable for identifying associations and testing a directional hypothesis in an applied student setting, but it does not establish causality with the same force as an experimental or longitudinal design. In addition, web-based surveys may be affected by self-selection and response-style issues, and non-probability sampling limits strict generalization to broader populations. These considerations do not invalidate the present results; rather, they define the scope of the study. Within that scope, the conclusion is clear: the uploaded evidence supports a positive and statistically significant effect of social media advertising on the purchase intention of Bina Nusantara University students, and the magnitude of the relationship is substantial enough to matter, both academically and practically.

This study examined the effect of social media advertising on the purchase intentions of Bina Nusantara University students in Jakarta, Indonesia. Using a quantitative design with 100 respondents and simple linear regression analysis, the results demonstrate that social media advertising has a statistically significant positive effect on purchase intention (t = 7.374, p < 0.001), explaining 35.7% of the variance in the dependent variable (R² = 0.357). The regression equation Y = 1.628 + 0.497X indicates a practically meaningful relationship, with each unit increase in social media advertising associated with nearly a half-unit increase in purchase intention.
These findings confirm that social media advertising is a significant behavioral influence on Indonesian university students and support its continued and expanded use as a digital marketing channel targeting millennial and Generation Z demographics. Marketing practitioners are encouraged to develop content that is visually compelling, contextually relevant, and socially credible while leveraging Instagram's integrated commerce features to minimize the friction between advertisement exposure and purchase action. Future research should extend the present model to incorporate additional predictors, platform-specific analyses, and actual purchase behavior as the outcome variable, with the aim of developing a more complete understanding of how digital advertising shapes consumer decision-making among young Indonesian consumers.