Mxit advertising ’ s in fl uence on cognitive attitudes amongst Millennials in the Western Cape

16The social media have grown at an exponential rate in recent years, especially amongst the youth (known as Millennials) in South Africa, who access social media primarily via mobile devices; these have served as an additional catalyst to fuel this growth. The rapid advancement of social media is also attributed to the Millennials’ desires for social interaction connectedness, information, entertainment and convenience. Millennials portray a diffi cult market to reach with advertising due to the large fragmentation of media, diverse range of interests and demographical differences, but this lucrative market cannot be ignored owing to their huge purchasing power. Mxit is a popular social medium in South Africa and provides numerous advertising opportunities for organisations and their brands. The purpose of this study was to examine Millennials’ cognitive attitudes towards advertising on Mxit, and assess the infl uence of certain usage and demographic factors. The research surveyed 1858 young adults in the Western Cape who were between the ages of 18 and 30 years. Millennials maintained favourable cognitive responses towards Mxit advertising. All of the usage elements and a demographic factor, gender, were also found to have a signifi cant infl uence. The results provide both academia and organisations with fresh insights and a greater understanding of social network advertising. 17

Mxit advertising's infl uence on cognitive attitudes amongst Millennials in the Western Cape R.G. Duffett 7 A B S T R A C T 16The social media have grown at an exponential rate in recent years, especially amongst the youth (known as Millennials) in South Africa, who access social media primarily via mobile devices; these have served as an additional catalyst to fuel this growth.The rapid advancement of social media is also attributed to the Millennials' desires for social interaction connectedness, information, entertainment and convenience.Millennials portray a diffi cult market to reach with advertising due to the large fragmentation of media, diverse range of interests and demographical differences, but this lucrative market cannot be ignored owing to their huge purchasing power.Mxit is a popular social medium in South Africa and provides numerous advertising opportunities for organisations and their brands.The purpose of this study was to examine Millennials' cognitive attitudes towards advertising on Mxit, and assess the infl uence of certain usage and demographic factors.The research surveyed 1858 young adults in the Western Cape who were between the ages of 18 and 30 years.Millennials maintained favourable cognitive responses towards Mxit advertising.All of the usage elements and a demographic factor, gender, were also found to have a signifi cant infl uence.The results provide both academia and organisations with fresh insights and a greater understanding of social network advertising.and marketing communication platform (Hoffman & Novak 2012: 69-70;Sobel 2010: 2).Social media have a major effect on society, since they have transformed previously passive consumers into digital advocates, especially among Millennials (the youth), who are willing to receive information about their favoured brands and disseminate it (Blasco-Arcas, Hernandez-Ortega & Jimenez-Martinez 2014: 226-229;Rajapat 2009: 28-90).Millennials are more technologically savvy than previous generations and constitute the main target market for online marketers (Okazaki 2009: 12-13;Strutton, Taylor & Thompson 2011: 560-561).Millennials are also the principal users of Mxit, which is the largest locally established social network and instant messaging conduit, with an estimated 4.9 million monthly users in Africa (Wronski & Goldstuck 2014).
The rapid increase in SNS across the globe and in South Africa makes social media very desirable for advertising.Advertising practitioners have long studied issues related to perception, attitudes and purchase intent toward brands.A substantial quantity of academic research has been conducted on advertising determinants and attitude responses on websites (De Haan, Wiesel & Pauwels 2013: 13-104;Lu, Chang & Yu 2013:27-68;Mina 2012: 209-222;Pi, Liao, Liu & Lee 2011: 7120-7129;Scarpi 2012: 53-67;Thales, Wedel & Pieters 2012: 144-159); however, these research studies have concentrated on traditional websites as opposed to SNS.Since social network advertising (SNA) is unlike traditional online advertising and much investigation is still necessary to find the true value of SNA, this study attempts to bridge this gap by providing research on attitudes towards SNA directed at Millennials (Saxena & Khanna 2013: 19;Weinberg, De Ruyter, Dellarocas, Buck & Keeling 2013: 308).Research on SNA, in particular Mxit, is virtually non-existent in the South African context (Kruger & Painter 2011: 49).Hence, this investigation intends to provide insight on Mxit advertising's perceived effect on awareness and knowledge, which is collectively referred to as the cognitive attitudinal stage (Belch & Belch 2015: 158) among Millennials in South Africa.

Literature background
Social media contextualisation 1In recent years, social media have redefined the digital landscape and altered the way that organisations disseminate marketing messages, particularly among Millennials.The social media are a platform that facilitates interaction between users around the world and allows them to share experiences with each other, as well as information that may also be brand related (Edwards 2011: 1 Heinrichs, Lim & Lim 2011: 347;Lipsman, Mudd, Rich & Bruich 2012: 41).Social media are different from traditional media and the Web, since they are generated by users who have control over the content generation (Dickey & Lewis 2011: 1-2;Peters, Chen, Kaplan, Ognibeni & Pauwels 2013: 282).This new platform has enabled marketers to use various forms of brand communication in an effective and interactive manner (Jothi, Neelamalar & Prasad 2011: 234).There has been a significant shift of marketing communication budgets in using digital media, with organisations investing in various SNA formats.Worldwide advertising via digital media was $120 billion in 2013 and is anticipated to increase to $204 billion by 2018, with the greatest growth coming from emerging markets (eMarketer 2014).SNS were not created to become advertising platforms, but rather with the objective of facilitating social interactions.It is therefore important to investigate users' attitudes towards social media marketing communications, since there are doubts regarding the effectiveness of SNA (Barreto 2013: 129-132;Weinberg et al. 2013: 308).

Mxit milieu
1Mxit is an affordable messaging platform that offers chat rooms, gaming, education platforms and community support services to its users, and operates mainly in the mobile environment (Kahn 2013;Stokes 2013: 475).Furthermore, Mxit provides companies with an opportunity to communicate with young consumers and to position promotions according to their budgets.A number of advertising options are available, which include text messages; content links; music and movie clips; branded content; sponsorships; apps creation; chat rooms, banner advertisements and splash screens (interstitials) that can be targeted according to specific demographics (such as age, gender and country) (Mxit 2014b).Another major advantage for brands is that Mxit reaches the lower to middle LSM groups owing to the fact that Mxit is inexpensive, easy to use and compatible with most mobile devices, which is essential in emerging markets where a majority of these markets do not own smart phones or computers (Britten 2010: 18;Petzer & Meyer 2013: 386).Furthermore, two-thirds of subscribers can be classified as Millennials, which makes Mxit a viable platform to reach this capricious market (Mxit 2014a).There is a dearth of inquiry into the effectiveness of Mxit as an advertising medium in terms of Millennials' attitudes towards this SNS.Hence, this study seeks to broaden academic discourse on this research topic.

Millennial cohort background
1Millennials are also referred to as Generation Y, Net Generation and Digital Natives, and in South Africa mostly comprise students, young working professionals and unemployed young adults (Prensky 2001: 1;McCrindle & Wolfinger 2009: 18-19;Pempek, Yermolayeva & Calvert 2009: 227-228).The exact years in which Millennials were born differ according to authors and countries, but the most commonly cited years are between 1982 and 1994 (Noble, Haytko & Phillips 2009: 617-628;Strutton et al. 2011: 560-561).One out of four South Africans is classified as belonging to the Millennial cohort (Statistics SA 2012: 23).This is the first cohort that grew up with modern technology such as computers, cell phones, smartphones, video games, iPods, iPads, tablets and SNS, which are intricately incorporated into their everyday lives to such an extent that they are addicted to technology for entertainment, interaction with others, and even for the regulation of emotions (Immordino-Yang, Christodoulou & Singh 2012: 352-364;Nikirk 2009: 20-23).Bevan-Dye and Dondolo (2014) reveal that South African students generally have favourable attitudes towards social media and that 86% of them access SNS such as Facebook via mobile phones.Levin (2013) estimates that this cohort spends more than R104 billion per year in South Africa (this only includes up to the age of 23 years, so the actual figure is much higher), which represents an important market for astute brands.Hence, Millennials' digital behaviour should be considered in order to obtain a better understanding of their online consumption practices and attitudes towards SNA and reach them more effectively with marketing communications.
Overview of hierarchy-of-effects model and attitudes 1Lavidge and Steiner's (1961: 59-62) hierarchy-of-effects model posits that advertising effects are a long-term process, in which consumers pass through successive stages, namely awareness, knowledge, liking, conviction, and desire to purchase before the final purchase.However, this model presumes that all consumers begin at the first hierarchy stage, whereas some consumers may have developed negative attitudes, which means that they would only progress once the unfavourable sentiment was removed.Consequently, the renowned terms 'cognition', 'affect' and 'conation' were first used in reference to the advertising hierarchy responses.Cognition (awareness and knowledge) refers to the rational states; affect (liking, preference and conviction) refers to emotional or feeling states; and conation (purchase intentions and purchase) refers to the striving or behavioural states (Barry 1987: 261-263;Belch & Belch 2015: 158).It is difficult to discuss consumers' perceptions and approval of SNA without taking into consideration their attitudes, since this aspect has been found to be an efficient measure of advertising effectiveness (Mir 2012: 271;Yoo, Kim & Stout 2010: 51).
Several studies have investigated attitudes towards advertising based on responses to SNA.Labrecque (2014: 137-139) reveals a positive cognitive (information) attitude regarding interaction with brands in social media environments.Hassan, Fatima, Akram, Abbas and Hasnain (2013: 327) established that consumers thought that SNA was informative.Chikandiwa, Contogiannis and Jembere (2013: 377-378) found that various SNS platforms were used successfully by South African banks to generate awareness and knowledge of their brands among consumers.Hollebeek, Glynn and Brodie (2014: 155-160) found that consumer brand engagement in social media had a positive influence on cognitive responses.Van Noort, Voorveld and Von Reijmersdal (2012: 229) posit that greater levels of online interactivity resulted in positive and stronger cognitive responses.However, a number of these inquiries were conducted in first-world countries, which may differ from developing countries; the samples comprised fewer than 400 respondents in most instances; students were predominantly utilised as the sample; and few considered the effect of independent variables such as demographic and usage factors on SNA.Hence, this inquiry seeks to provide a deeper understanding of the cognitive attitudinal responses of South African Millennials to advertising on Mxit.

Problem statement and objectives
1The hierarchy-of-effects model stages are analogous to the purchase funnel, in which consumers' progress through a number of stages until the ultimate purchase, but it becomes progressively more difficult to reach higher level stages (Belch & Belch 2015: 228;Safko 2010: 91-93).Yet, marketing communication notional models were developed via traditional above-the-line advertising, while this investigation focuses on the most recent online social media to determine consumers' cognitive responses as they move through the first two lower hierarchy levels (Barry 1987: 251-295).Consequently, this empirical research aims to determine the impact of Mxit advertising on the cognitive responses of Millennials in South Africa.Wronski and Goldstuck (2013: 116) reveal that nearly 95% of Mxit users access this social medium via mobile phones; hence, it would be significant to ascertain if there was a difference between Millennials' attitudes in terms of how they accessed Mxit.Usage period, incidence of log on, log on duration and profile update rate are other usage characteristic factors that were examined to establish if they have an effect on Millennials' cognitive attitudes towards advertising on Mxit.
Bolton, Parasuraman, Hoefnagels, Migchels, Kabadayi, Gruber, Loureiro and Solnet (2013: 257) reveal that there is a dearth of social media research to determine whether there are noticeable differences within the Millennial subgroups.Wang, Yu and Wei (2012: 206) advise that more social media research should be conducted in other countries, since consumer sentiments from dissimilar cultural backgrounds would differ.Hence, this inquiry seeks to ascertain whether demographic factors such as age, gender and ethnicity have an influence on Millennials' cognitive attitudes towards SNA.
Therefore, the research objectives of this study can be summarised as follows: • To ascertain the impact of Millennials' cognitive attitudes towards Mxit advertising.• To determine whether usage characteristics affect Millennials' cognitive attitudes regarding Mxit advertising.• To establish whether Millennials' demographic variables have an effect on cognitive attitudes towards Mxit advertising.

Methodology
1In this inquiry, the research design is descriptive and employs the survey method.The problem and factors such as individuals' attitudes towards an object (for example, Millennials attitudes towards Mxit advertising) constitute descriptive research, which is utilised to describe what is being researched (Churchill & Lacobucci 2004: 91).This form of research is characteristically employed when knowledge of the object that is studied is nebulous or vague (Wiid & Diggines 2009: 55), and not much is known about attitudes towards Mxit advertising in South Africa.Descriptive research usually takes a cross section of a population (in the form of a survey); in this instance, Millennials that live in the Western Cape, and reveals their predisposition at a given point in time on which the research can be built.

Sampling
1Several inquiries have identified Millennials as the predominant users of online social media and mobile conduits, and accordingly a potentially lucrative target market for astute marketers (Grant & O'Donohoe 2007: 223-246, Logan, Bright & Gangadharbatla 2013: 164-179;Taylor, Lewin & Strutton 2011: 258-275).Other studies have found that students are the main SNS users (Jordaan, Ehlers & Grove 2011: 1-20, Saxena & Khanna 2013: 17-25) and are, therefore, susceptible to SNS advertising, which is used on platforms such as Mxit.It was essential to draw a representative sample of Millennials other than exclusively using students, who were the sole focus of many other social media-related studies (Bannister, Kiefer & Nellums 2013: 1-20;Barreto 2013: 119-139;Chandra, Goswami & Chouhan 2012: 1-14;Mir 2012: 265-288;Ruane & Wallace 2013: 315-335).Accordingly, young adults who had matriculated and who were either unemployed or employed and studying were included in the research population to ensure that an all-inclusive representation of Millennial Mxit users in the Western Cape was accomplished.A multi-stage sampling technique (probability sampling) was adopted (Bhattacherjee 2012: 68-69) in which the Cape Peninsula was divided into various geographic areas (clusters) that incorporated both suburbs and informal settlements.Thereafter, a range of community groups, tertiary education institutions and businesses (sample frame) were randomly targeted to participate in the study.These organisations were contacted telephonically to obtain the necessary approval to conduct a survey on their premises.Furthermore, permission (and ethical clearance) was sought and obtained from the Western Cape Education Department to conduct research in schools and colleges, and approval (and ethical clearance) was also received from the Cape Peninsula University of Technology.Systematic sampling was then employed to survey respondents at the aforementioned organisations, with every third respondent being individually approached and invited to voluntarily participate in the study by the fieldworker.However, each potential respondent was first qualified as a member of the Millennial cohort and as a Mxit user (sample element) who had observed advertising on this platform.The respondents were informed that they had the right to: privacy (i.e. could refuse to participate in the study or to answer any question); anonymity and confidentiality (no personal details were requested, i.e. name and/or contact particulars); and full disclosure (a full explanation of the research was communicated to all respondents).

Questionnaire design, data collection and analysis
1A self-administered research instrument was employed, since it permitted more rapid collection of data than face-to-face interviews, but still resulted in a high response rate.The participants completed the questionnaires, while fieldworkers were available to provide assistance if required (De Vos, Strydom, Fouché & Delport 2011: 188-189).The first section of the research instrument comprised multiplechoice questions focusing on five social media Mxit usage factors (independent variables).The second section explored the dependent variables, awareness and knowledge (cognitive attitude component) by means of two nine-item scales that were largely developed from Ducoffe (1996: 21-35) and Duncan et al. (1985: 33-64).A five-point Likert scale was used to gauge responses, which ranged from strongly disagree to strongly agree (Cant, Gerber-Nel, Nel, & Kotzé 2008: 141).The statements used to measure the awareness and knowledge constructs are listed in Table 1.Likert scale items were arranged so as to systematically switch between negatively and positively phrased statements to avoid response bias.The negatively phrased items were first reversed before Cronbach's alpha was employed to establish construct reliability (Pallant 2010: 85-88).Cronbach's α was 0.831 for the Mxit advertising awareness scale and 0.810 for the Mxit advertising knowledge scale, which displayed reputable internal consistencies.Finally, multiple-choice questions established three demographic characteristics (independent variables).A pilot study was employed among 50 participants to determine the construct reliability; confirm that the multiple-choice questions were adequate; and ensure that the entire research process was feasible and efficient (De Vos et al. 2011: 236-246).The data were coded, captured and analysed using SPSS (version 22) statistical software.A majority of the respondents (73.4%) accessed Mxit only via mobile devices, routinely logged on to Mxit every day (64.4%),spent one (29.8%) to two (25.0%)hours per log in, and updated their profile at least once a week (68.3%).The sample was slightly dominated by females (52.5%); a majority of the respondents were 18-20 years old (62.5%), which positively correlates with Mxit's audience demographics (Mxit 2014a); and the ethnic orientation reflected population groups that reside in the Western Cape with Black (51.8%) and Coloured (38.3%) being majorities (Statistics SA 2012: 17).

Cognitive attitudes 1The
Wald's chi-square test displayed a significant difference at p < 0.001 for awareness (M = 3.53, SD = 0.787) as a consequence of exposure to Mxit advertising.Mxit advertising indubitably caused favourable cognitive responses among Millennials in South Africa, thereby realising the first research objective.Hutter, Hautz, Dennhardt and Füller et al. (2013: 347) reveal that a consumer's interaction with a company's SNS page has a positive effect on the consumer's awareness of the company brand, which is in congruence with the results of this study.Schivinski and Dąbrowski (2013: 14) confirm that organisations' SNS communications create awareness and favourable associations with their brands.SNA promotes greater awareness of brands among consumers, and plays an important role in influencing buying decisions for online purchases (Agrawal & Jaliwani 2013: 135).
The Wald's chi-square test proved that there was a significant difference at p < 0.001 for knowledge (M = 3.45, SD = 0.773) as a result of Mxit advertising.One of companies' main reasons for using SNA is to supply information about their brand, products and services to their consumers; whereas many consumers have stated that the key reason why they accept advertising is owing to company information (Saxena & Khanna 2013: 18).Manomayangkul (2012: 83-86) also determined that product information positively affects consumer attitudes towards SNA.The proliferation of the internet and social media has given access to more consumption-related information than any previous generation, including online advertising.The capability of SNS to collect details about consumers in terms of purchase, consumption cycles and search history may assist marketers to regain some power they have lost to consumers since the advent of the interactive digital online environment (Labrecque, Vor dem Esche, Mathwick, Novak & Hofacker 2013: 262).
Usage characteristics 1The significant differences between the dependent (awareness and knowledge) and independent (usage and demographic factors) were established by employing a generalised linear model (GLM) analysis of variance (ANOVA), and utilising Wald's chi-square and the post ad hoc Bonferroni adjustment pairwise comparisons (Pallant 2010: 283-296) to detect whether there was a significant difference among the abovementioned variables.The Bonferroni correction pairwise comparisons of estimated marginal means found significant differences for a majority of the usage characteristics displayed in Table 3.  findings with respect to each independent variable are discussed: • Access (p < 0.001): Mxit accessed via a mobile device (M = 3.48, SE = 0.043) displayed higher awareness levels than when accessed by personal computer (M = 3.18, SE = 0.091).Mxit is a predominantly mobile SNS, hence it is reasonable to accept that advertising would be more effective on mobile devices than on desktop computers.An advantage of Mxit is that it can be accessed on almost any mobile device, including smartphones, tablets, feature phones and other cell phones.Millennials who had utilised Mxit for an extended period of time displayed more favourable cognitive responses than those who had utilised the social medium for a shorter period.This is not an unexpected result, because consumers have become more familiar and comfortable with this mobile SNS platform and are more likely to become aware of advertisements that appeal to them since they are target-market specific.• Length of usage (p < 0.001): Respondents who had used Mxit for 5 years (M = 3.52, SE = 0.056) showed increased levels of awareness compared to those who had used the SNS for 1 year (M = 3.21, SE = 0.066), 2 years (M = 3.28, SE = 0.059), 3 years (M = 3.35, SE = 0.057) or 4 years (M = 3.36, SE = 0.059).
Respondents who had used Mxit for 5 years (M = 3.45, SE = 0.055) also displayed greater levels of knowledge than those who had used the SNS for 1 year (M = 3.24, SE = 0.065), 2 years (M = 3.21, SE = 0.058) or 3 years (M = 3.29, SE = 0.057).Several inquiries confirmed that the users' experience levels in terms of the internet had an effect on the consumer decision-making process (Helander & Khalid 2000: 609-619;Liao & Cheung 2001: 299-306;Montoya-Weis, Voss & Grewall 2003: 448-458).• Log on frequency (p < 0.05): Mxit users who logged on daily (M = 3.46, SE = 0.047) yielded higher awareness levels than those who logged on once a week (M = 3.26, SE = 0.069).Mxit users who logged on daily (M = 3.42, SE = 0.046) also yielded higher knowledge levels than those who logged on once a week (M = 3.20, SE = 0.068).Sago (2013: 8) also found that respondents who accessed the internet more regularly generally spent increased amounts of time online, which resulted in greater exposure to marketing information and interest in online advertisements.Millennials who logged on regularly resulted in greater positive cognitive responses than those who logged on less frequently, which is a fair supposition, as the more times Mxit users log on (the average is 5 times a day), the greater the opportunity they have of becoming aware of and engaging with SNA (Wronski & Goldstuck 2014).• Log on duration (p < 0.05): Respondents who logged on for 1 hour (M = 3.32, SE = 0.52) yielded increased knowledge levels compared to those who remained logged in for 5 hours (M = 3.15, SE = 0.061).The analysis showed that Millennials who spent an hour or less logged on to Mxit had more favourable attitudes to knowledge.One would expect the converse to be true, since the longer consumers spent on this mobile platform, the more time they would have to view advertising and gather information on brands.However, Millennials generally multitask and access Mxit and other SNS whenever they have a moment, for example, while waiting in an queue or travelling, hence they commonly do not have time to spend extended periods of time per log on (Crux Research 2013; Du Chenne 2011: 50; Smith 2012: 86-92).However, this investigation confirmed that over half of Millennials spend two hours on Mxit per log on.• Profile update incidence (p < 0.05): Respondents who updated their Mxit status daily (M = 3.44, SE = 0.058) showed greater awareness than those who updated once a month (M = 3.27, SE = 0.057).Respondents who updated their Mxit status 2 to 4 times a week (M = 3.34, SE = 0.058) and once a week (M = 3.38, SE = 0.057) also exhibited greater knowledge levels than those who updated it once a month (M = 3.18, SE = 0.056).Millennials who updated their Mxit profile more frequently display positive cognitive responses to advertising on this conduit.This finding is rational, as the greater interaction on the SNS would lead to the increased likelihood of engagement with other Mxit marketing communication elements.
Demographic variables 1No significant differences were reflected for age and ethnic orientation in terms of awareness, nor for any of the demographic variables in terms of knowledge.However, Bonferroni correction pairwise comparisons of estimated marginal means revealed a significant difference for gender vis-à-vis awareness, as shown in Table 4. Wald's chi-square test showed a signifi cant diff erence at p<0.05.
1As regards the findings with respect gender (p < 0.05), males (M = 3.39, SE = 0.049) displayed higher awareness levels than females (M = 3.30, SE = 0.049).A number of studies determined that there was no difference in terms of gender in attitudes towards SNA (Agrawal & Jaliwani 2013: 136;Barreto 2013: 23); however, this investigation found that males tended to have a more favourable predisposition regarding awareness of Mxit advertising.Sago (2013: 8) confirms that gender differences exist among the Millennial cohort in terms of the use of social media.Females were found to have lower levels of satisfaction from using SNS than males.Davis and Rosenstein (2012: 4) also report that men notice SNA and click on advertisements more frequently than women.However, women have progressively begun to use SNS; they have increased spending power and are more willing to make purchases online than men (Walter 2014); therefore, organisations cannot afford to ignore women in their SNA.

Theoretical implications
1A dvertising on Mxit has a positive influence on cognitive attitudinal responses among Millennials, which authenticates the essentials of the hierarchy-of-effects model that was created by means of above-the-line advertising.Consequently, it can be posited that Mxit advertising abides by the same principles of traditional advertising, since it results in comparable awareness and knowledge sentiments.The research outcomes are also analogous to inquiries in first-world nations, thus validating that parallel cognitive attitudes are evident between developing and developed countries (Hollebeek et al. 2014: 155-160;Hutter et al. 2013: 346-347;Labrecque 2014: 137-139;Schivinski & Dąbrowski 2013: 14;Van Noort et al. 2012: 229).Furthermore, the investigation also demonstrates that Mxit advertising causes corresponding cognitive responses to internet advertising in first-world nations (Chatterjee 2008: 51-61;Yoo et al. 2010: 55).Bolton et al. (2013: 251-257) propose that globalisation may lead to increasing homogeneity among Millennials in terms of SNS usage.However, owing to the large disparity between wealth, infrastructure and technology among emerging and developed countries, there has been discourse regarding the accuracy of this viewpoint (Petzer & Meyer 2013: 382-386).Several of the studies also predominantly investigated Millennials (Hutter et al. 2013: 346;Schivinski & Dąbrowski 2013: 9-10), and a comparison of the results of this study with those investigations shows that Millennials share equivalent SNS usage and attitudes in South Africa along with first-world countries, proving the assumption of Bolton et al. (2013: 251-257) to be correct.Hence, important additions have been made to advance the hierarchy-of-effects model and generational research in terms of SNA.Bolton et al. (2013: 257) prescribe that investigations should also ascertain whether there are noticeable differences within Millennial subgroups.This analysis explored whether particular usage factors and demographic variables affect Millennials' awareness, and verified that a majority of the usage factors affect Millennials' cognitive responses as a result of Mxit advertising.Few demographic variables were found to have an impact on Millennials' awareness and knowledge, but gender was shown to influence affective responses towards advertising on Mxit.Hence, noticeable differences were revealed in Millennial subgroups, and thus this innovative research has contributed to the body of knowledge on attitudes to SNA and cohort theory.

Practical implications
1The favourable cognitive responses towards Mxit advertising, as shown in this study, reinforce the significance of SNS marketing communications.Though some marketers may still doubt the effectiveness of SNA, this research shows that SNS are integral and essential to promotional communication strategies.The favourable influence of Mxit advertising on awareness and knowledge is a strong incentive in terms of the relevance of including SNS in integrated marketing communications campaigns.Barreto (2013: 132) also emphasises that awareness of the full potential of advertising via SNS would fast-track organisations' ability to reach their consumers more efficiently.RedStage (2014) discloses that Millennials prefer more detailed brand information, consumer reviews and social interaction when shopping online; and Redsicker (2013) reports that almost seven out of ten consumers search for information about brands on social media.This represents a valuable opportunity for marketers to inform consumers by means of convincing, indepth and educational content.However, the information provided by organisations should be credible, relevant, complete, truthful, recent and timely, as this will assist in decreasing the time required to find and examine brands, thereby reducing doubt when making consumer purchasing decisions (Bevan-Dye & Dondolo 2014).Mxit is a wide-reaching and engaging platform that provides a number of interactive mechanisms to increase cognitive responses towards brands via content links, chat rooms, text messages, downloads, apps and many more promotion tools by providing information.
A number of usage characteristics were established to be an effective recipe for the high cognitive responses towards advertising on Mxit among Millennials.A majority of individuals tend to have their mobile devices with them on a 24/7 basis, which implies that they are always available and connected.This in turn provides marketers with significant marketing communication opportunities to reach and connect with their target audiences quickly and easily (Barker et al. 2013: 264).There are a number of entertaining and interactive Mxit branded apps such as Tradepost, which longtime users become accustomed to over time, ultimately leading to increased cognitive responses.Furthermore, Mxit provides a multimedia-enabled environment that allows for multi-group dialogue between consumers and brands (Mxit 2014b).Mxit users who log on daily, spend less than one hour and update their profiles regularly also displayed high awareness levels and knowledge perceptions towards advertising; therefore, organisations should contemplate a number of branded communication possibilities to encourage Millennials to log on multiple times a day, for example, via games, branded apps, chat rooms and splash screens.
Marketers still persist in spending more of their advertising budget in targeting men on SNS (Davis & Rosenstein 2012: 5), which is in line with the results of this analysis, as women were shown to have less positive awareness sentiments.However, De Lanerolle (2012: 8-9) verified that the gender gap was beginning to diminish in the online digital environment.RedStage (2014) discloses that women spend 20% more time than men viewing internet shopping sites and also spend more time in terms of online purchases than men.Males are more task-orientated when shopping online and are inclined to search for more information about a specific product, while women are inclined to browse different products and only scan information, leading to a greater frequency of impulse purchases.Women are also more dependent on SNS when making purchasing decisions.Therefore, although the findings revealed that men have marginally higher awareness perceptions towards Mxit advertising, organisations should not disregard women in their marketing communications on SNS, especially since South African consumer purchases via online platforms are expected to triple over the next five years to almost R60 billion (Wilson 2013).
Limitations and future research 1No differentiation was made between the several types of Mxit advertising, which were jointly investigated, but further studies could analyse these individually to explore whether they yielded different results.Millennials were the focus of the inquiry, though other generational cohorts, Generation Z and Baby Boomers, could also be analysed in a comparative study to provide more inclusive insight into perceptions of advertising on Mxit.The survey produced a snapshot of awareness and knowledge perceptions, whereas a longitudinal research design could be employed to provide a thorough portrayal over a prolonged time period.A qualitative inquiry would yield greater insight into Millennials' attitudinal responses, as would an analysis of affective and conative attitudes.Future research could take into consideration other provinces (apart from the Western Cape) to obtain a more complete representation of South Africa owing to possible regional differences.This study could be repeated for other significant SNS (e.g.Twitter, YouTube, LinkedIn, Google+ and Facebook) and in other emerging and first-world countries to ascertain whether there are noteworthy differences and hence attain a more far-reaching grasp of SNA.

Conclusion
1The revolution on social media has forever changed the marketing communication milieu.The increased prominence of social media platforms in consumers' lives has a growing impact on their communication behaviour.Millennials are spending increasing amounts of time on SNS as an increasing share of advertising moves to this new social media landscape.This interactive climate promotes brand engagement and exposure to marketing communication, which is rapidly increasing within SNS.This new phenomenon creates new opportunities and threats for companies' marketing communications, as purchasing decisions are progressively affected by SNS engagement (Hutter et al. 2013: 343).This investigation provided new theoretical and practical knowledge regarding Mxit advertising by establishing that Millennials, the main users of this mobile SNS, predominantly maintain positive cognitive responses.The study confirmed that a number of usage characteristics impacted on the awareness and knowledge perceptions of Mxit advertising, as did the demographic variable of gender.Therefore, this research is significant in academic circles, as it has provided additional information that will aid the expansion of the attitudinal, cohort and social media advertising discourse.The number of young adults who use SNS such as Mxit to assist with purchasing decisions has grown exponentially, and millions are anticipated to visit e-buying outlets (Agrawal & Jaliwani 2013: 135), hence this study has also provided corporate decision-makers with further insight into attitudes towards SNA among Millennials in the Western Cape to aid with their digital marketing communication strategies. 1The 1Atotal of 1858 members of the Millennial cohort participated in the Mxit survey in the Western Cape. 1The

Table 1 :
Mxit advertising awareness and knowledge scales mcxliiiA wareness mcxlivA dvertisements on Mxit are eff ective in creating awareness of brands mcxlvA dvertising on Mxit alerts me to new company off erings mcxlviI have NOT become aware of new advertisements on Mxit* mcxlviiA dvertisements on Mxit attract my attention to certain brands dvertising on Mxit does NOT alert me to new products* dvertisements on Mxit do NOT inform me about a company's off erings* mclvA dvertising on Mxit is a convenient source of information mclviA dvertisements on Mxit provide me with new product knowledge mclviiA dvertisements on Mxit do NOT provide much information on brands* mclviiiA dvertisements on Mxit provide useful data about company off erings mclixA dvertisements on Mxit are eff ective in providing information on brands mclxA dvertisements on Mxit are a POOR source of knowledge* mclxiI use advertisements on Mxit to fi nd new information about products mclxiiA dvertisements on Mxit provide me with valuable product knowledge mcliiA mcliiiKno wledge mclivA Table 2 provides a comprehensive summary of Mxit usage characteristics and of the Millennial participants' demographics.

Table 2 :
Mxit usage characteristics and demographic variables

Table 3 :
Eff ect of usage characteristics on Mxit advertising cognitive attitudes

Table 4 :
Eff ect of demographic variables on Mxit advertising cognitive attitudes