The Attitude of Generation Z Towards YouTube Skippable Ads: An Empirical Study on Lokalate Ads

As businesses’ marketing activities have migrated to various social media platforms, competition between advertisers to attain market’s attention is getting more intense. As a result, marketers began to involve manipulative and provocative strategies to steal market’s attention. This, however, lead to a boomerang effect where consumers develop negative attitude towards advertisement—especially among Generation Z as one of the biggest demographic groups. Aimed to fill in the knowledge gap regarding this issue, specifically in Medan, this research investigates how attitude towards advertisement is affected by several antecedents, namely informativeness, entertainment, and irritation. Multiple linear regression analysis is employed to assess the relationship between variables. The result shows that informativeness along with entertainment have positive effect on attitude towards advertisement, whereas irritation has negative effect. This indicates that marketers should consider these factors to create an effective advertisement that can evoke positive attitude among audiences.


INTRODUCTION
The world is experiencing swift digitalization evolution, given the emergence of various social media platforms. This explains why businesses began to embrace online and social media channels to better reach their market. In Indonesia alone, the number of social media users increased from 160 million as of January 2020 (Kemp, 2020) to become 170 million users by January 2021 . Compared to neighbouring Southeast Asian countries, this figure in indeed the highest. In addition to the massive number of audiences, social media platforms are equipped with advanced technological features that allows business marketers to significantly save cost-both monetary and non-monetary.
Among a list of rising social media platforms in Indonesia, Data Reportal (2021) declares that YouTube is on top of the list with the highest penetration rate of 93.8% out of Indonesian social media users aged 16-64. In return, YouTube has become one of the most popular channels utilized by businesses to communicate with their market. Aside from its massive audience, YouTube also offers marketers with extensive variety of advertising format, namely: (1) skippable video ads, (2) non-skippable video ads, (3) bumper video ads, (4) display ads, along with (5) overlay ads. According to a worldwide survey among marketing professionals conducted by Statista (2019) during mid-2019, YouTube 15-seconds skippable video ads played before (pre-roll) a video content were considered the most effective YouTube ad format. This explains why YouTube has become one of the most opted channels by advertisers in their marketing strategy.
As marketers are migrating to the world of social media, however, competition for consumer's attention is getting more intense (Sim, 2016). This phenomenon pushes advertisers to come up with new and different approaches in order to stand out and get recognized by their audiences. Some decided to intensify their ads exposure (Sabuncuoğ lu-İ nanç et al., 2020), while others decided to involve exaggerated, misleading (Sim, 2016), or even daring advertising contents (Ferreira et al., 2017). In fact, these efforts bring more harm than good to the business since they majorly contribute to the shaping of negative consumer attitude towards advertisement.
One manifestation of this attitude is revealed by a study, claiming that 59% of millennials is willing to watch a YouTube video ad only until the skip option is available (Kirkpatrick, 2017). This concerning phenomenon is supported by the findings of Extreme Reach (2020) which concluded that there is a drop in worldwide average advertisement Video Completion Rate (VCR) from 89% in 2019 to 80% in 2020. A segmented study focusing on Generation X, Y and Z conducted by Kantar Millward Brown (Sterling, 2017) also lends a support to this issue, in which it is discovered that Generation Z has the most negative attitude towards various forms of advertisement compared to the other two age group. Given that Generation Z has an uncanny population size-2.47 billion individuals in 2019 (Spitznagel, 2020), marketers should pay attention to and understand the needs along with attitudes of this specific age segment.
To understand consumer's attitude towards YouTube ads-particularly skippable video ads, marketers should first be aware that attitude is shaped by perception towards the advertisement value itself. This is because how audiences respond and act towards an advertisement will depend on their cognition (thinking) and affection (feeling) towards it. Some antecedents that affect attitude towards ads includes informativeness, entertainment and irritation (Adidarma & Putri, 2017;Qousa & Wadhy, 2018;Sabuncuoğ lu-İ nanç et al., 2020;Wiliyastuti et al., 2018). Informativeness concerns consumer's cognition which assesses if an advertisement is supplied with relevant knowledge regarding the advertised products or services. While entertainment and irritation relate to affection. Entertainment elicits positive feeling, where it evaluates the enjoyment provided by an ad. In contrary, irritation trigger negative feelings when an advertisement exhibit displeasure to the audience. Thereby, an ideal ad is one that is perceived to be informative, entertaining, and not irritating The phenomenon in which Generation Z, as one of the largest demographic groups, exhibit the most negative attitude towards advertisement is not something that can be overlooked. Reason is because consumer's attitude towards advertisement majorly determines whether company's message has been successfully communicated to the market or not. This is especially true for Lokalate-Nutrifood's instant coffee brand-who focuses on targeting the younger market and is utilizing YouTube as their marketing channel. Despite the critical issues, studies that examine this topic is inadequate. This serves as an urgency for this research to answer the knowledge gap regarding Generation Z attitude towards YouTube skippable ads. Hence, this study was carried out to investigate the influence of informativeness, entertainment and irritation towards Generation Z attitude towards Lokalate's YouTube skippable ad. Given that none of the previous studies explore this topic specifically among Generation Z in Medan, this research will focus and provide insights in relation to the mentioned segment. This study provides empirical research and managerial implication for marketers in Medan to optimize their marketing strategies.

Attitude towards Advertisement
The extent to which an advertisement is informative, entertaining, and irritating will affect consumer's attitude towards it. This is because attitude generally describes an individual's consistent cognition, affection and conation towards an object or idea that could either be favourable or unfavourable (Kotler & Armstrong (2017: 172). This is consistent to how MacKenzie & Lutz in Boon et al. (2019) elaborate attitude, which is the inclination to respond pleasantly or unpleasantly given a stimulus. Since attitude assesses how one will think, feel, and likely react towards a particular thing, attitude internally determines one's behaviour in a significant manner. Attitude is the consistent inclination which will shape behaviour as the outcome. In this research, attitude towards advertisement is evaluated through perception, feeling or emotions and behaviour.

Informativeness
Informativeness is a quality of an advertisement which indicates that it supplies sufficient information for consumers to make better purchasing decision (Alalwan, 2018;Wiliyastuti et al., 2018). Since it is associated to cognition, an advertisement should be equipped with logical knowledge to satisfy audiences' rational thinking and, thus, rationally convince them to make a purchase. Informative advertisement is characterized with being a good source of product information, providing relevant knowledge regarding the product and delivering up to date information.

Entertainment
Entertainment is linked to affection and emotion of the audiences. This antecedent addresses the degree to which an advertisement can bring enjoyment, relaxation, and fun to grab and maintain audiences' attention towards the ad (Hidayat et al., 2016;Setiawan & Briliana, 2021). Some properties that assess an advertisement's entertainment includes its ability to be entertaining, attractive, and enjoyable for the audiences.

Irritation
Similar to entertainment, irritation also concerns emotions and feelings but negatively. Irritation can be inferred as the capacity in which an advertisement exhibits annoyance, disturbance, interruption and displeasure (Gaber et al., 2019;Hidayat et al., 2016;Wiliyastuti et al., 2018). Irritation can be measured by the level of annoyance, deception and confusion audiences experience as a result of watching certain advertisement.

Informativeness and Attitude towards Advertisement
Informative advertisement provides consumers with tangible benefits (i.e., knowledge, information, understanding) which allows consumers to make better decision (Tjahjadi & Hendijani, 2020). As informativeness adds value to advertisement and allows watching advertisement a worthwhile experience, more positive attitude can be established. This is consistent to previous studies stating that there is a positive relationship between informativeness and consumer's attitude (Adidarma & Putri, 2017;Boon et al., 2019;Puwandi et al., 2020;Wiliyastuti et al., 2018).

H1: Informativeness of Lokalate YouTube skippable ads affects Generation Z attitude Entertainment and Attitude towards Advertisement
Advertisement perceived to be entertaining create an enjoyable and good experience for the audience. This explains why entertainment is believed to boost the value of an advertisement (Puwandi et al., 2020). In contrary, advertisement perceived to be boring will be ignored as it is deemed to be a waste of time. In conclusion, entertainment influences attitude positively as illustrated by previous literatures (Adidarma & Putri, 2017;Boon et al., 2019;Qousa & Wady, 2018;Sabuncuoğ lu-İ nanç et al., 2020;Wiliyastuti et al., 2018).

H2: Entertainment of Lokalate YouTube skippable ads affects Generation Z attitude Irritation and Attitude towards Advertisement
People will naturally avoid what irritates them; this goes the same with advertisements. This clearly means that irritating advertisement will elicit negative associations and, thus, negative attitude. The fact that irritation can affect attitude negatively is supported by several prior researches (Chungviwatanant et al., 2017;Qousa & Wady, 2018;Wiiliyastuti et al., 2018).

H3: Irritation of Lokalate YouTube skippable ads affects Generation Z attitude
Based on the above discussions, the research model below is proposed:

METHODOLOGY
This research was conducted by referring to one of Lokalate's YouTube video ads titled: "BUY 1 GET 1: LOKALATE KOPI ALPUKAT!". The population involved is potential Lokalate consumers (i.e., those who had never consumed any Lokalate's product before) who are Generation Z and domiciled in Medan. Given the unknown population, non-probability sampling was conducted, specifically purposive sampling.
The rule of thumb suggested by Hair et al. (2019: 132) mentioned to include a sample size that is equivalent to five or 10 times as many items used to measure a variable. The total items used to measure four variables in this study is 24. Thereby, the minimum sample size to yield a good result will be 24 x 5, which is equal to 120 samples. In this study, the sample size employed will be 300 samples.
Primary data used in this research are collected through online questionnaire comprising of 24 questions. The online questionnaire is distributed to individuals who meet the criteria set by researcher, namely: (1) born between 1996-2010 (Generation Z), (2) had never consumed Lokalate's product before, (3) domiciled in Medan. Before being distributed, the questionnaire has gone through a pre-test on 30 respondents to check its validity and reliability. The validity test is done by using the Pearson formula as shown below, where rxy denotes the correlation coefficient of x and y, n denotes the number of samples, ∑xy denotes the sum of x multiplied by y, ∑x denotes the sum of x, and ∑y denotes the sum of y: For the reliability of instrument, it is tested by using the Cronbach Alpha with the formula as follows, where ra is the reliability coefficient of Cronbach's Alpha, n is the number of questions and t 2 is the total variance of each question: In analyzing the relationship between independent variables (i.e., informativeness, entertainment, irritation) and dependent variable (i.e., attitude towards advertisement), the researcher adopts multiple linear regression with the help of SPSS 26.0 software as the data analysis technique. Before jumping to multiple linear regression, four classical assumption tests should be carried out.
To begin with, normality of data distribution should be tested. Normality test can be achieved by utilizing Kolmogorov-Smirnov non-parametric with the help of SPSS software. Upon calculation, data with significance level (α) > 0.05 is considered to follow normal distribution. While data with significance level (α) < 0.05 does not follow normal distribution (Sugiyono, 2020:234).
After so, linearity test is carried out to assess if relationships between independent and dependent variables are linear or not. The test involves the fulfillment of certain criteria for the relationship between independent and dependent variables to be deemed as linear. If the value of significant deviation from linearity falls ≥ 0.05, the relationship is linear, while significant deviation from linearity that falls < 0.05 indicates non-linear relationship. The third test is multicollinearity aimed to evaluate relationship between independent variables. A good regression model is when independent variables show no correlation. Multicollinearity can be identified through Value Inflation Factor (VIF) and tolerance value. Similar to the previous test, multicollinearity test involves the fulfillment of certain criteria. VIF > 10 with tolerance value < 0.1 demonstrates the presence of multicollinearity, whereas VIF < 10 with tolerance value > 0.1 shows the absence of multicollinearity (Ghozali, 2018:104).
Lastly is to carry out heteroscedasticity test, a test to examine the presence of disparity in residual variance within observations. Checking the heteroscedasticity can be done by employing the Park test that is available in SPSS software. When the value of significance < 0.05 it can be learnt that heteroscedasticity happens in the model. While significance value > 0.05 suggests that there is no heteroscedasticity. The formula of Park test is (Ghozali, 2018:136): As the classical assumption tests are met and satisfied, data multiple regression analysis can then take place. This analysis is aimed to predict and examine the influence of independent variables on the dependent variable. This relationship can be evaluated by using the following formula: Indonesian Journal of Business Analytics (IJBA) Vol. 1 No. 1, 2021: 17-32 Y explains the dependent variable, α explains the constant of y-intercept, β1 explains regression coefficient of the first independent variable, β2 explains regression coefficient of the second independent variable, β3 explains regression coefficient of the third independent variable, X1 explains the first independent variable (i.e., informativeness), X2 explains the second independent variable (i.e., entertainment) and X3 explains the third independent variable (i.e., irritation).
Determination test is then conducted to measure the strength in which an independent variable can describe the variations of dependent variable. Determination coefficient is denoted as R 2 that may fall in between 0 to 1. R 2 closer to 1 implies that the independent variable can explain nearly all the information required to estimate the changes in dependent variable. The following is the formula for determination test, in which R 2 is the coefficient of determination, n is the number of samples, ∑xy is the sum of x multiplied by y, ∑x is the sum of x, and ∑y is the sum of y: To statistically test the accuracy hypothesis proposed, this research will utilize t-test. The t-test formula used is shown below: From the formula above, t represents the value of t-test, r represents the correlation between independent and dependent variables, while n represents the number of samples. After calculating the value of t-test, it will be compared to the value of t table. When the value of tcounted ≥ ttable or -tcounted ≤ -ttable, the hypothesis will be accepted and vice versa. Decision can also be done by analyzing the significance probability value (p-value), in which p-value < 0.05 signifies acceptance and p-value ≥ 0.05 signifies rejection of hypothesis.

Validity Test
The calculation of each measurement item's r value to test the validity of instrument is done with the help of SPSS 26.0. All six items measuring Informativeness (INF) is considered valid, as shown in the following table:  (2021) The variable Entertainment (ENT) consists of six measurement items which is computed to be valid according to the r values elaborated in the table below:  (2021) The r value of all six items measuring Irritation (IRR) is shown in the following table in which all are considered to be valid:  (2021) Similarly, all six items measuring attitude towards advertisement are considered valid with r values detailed in Table 4 as follows:

Reliability Test
The value of Cronbach's Alpha for reach variable is obtained with SPSS 26.0 software. Reliability is achieved when the value of exceeds 0.7. The following table shows the Cronbach's Alpha value of the first variable, Informativeness (INF), which is seen to be reliable: Cronbach's Alpha value for Entertainment is shown in Table 6 below, indicating that the items measuring the variable is reliable:  (2021) The table below shows the Cronbach's Alpha value for Attitude towards Advertisement (ATA) and the value suggests that the items appear to be reliable:

Normality Test
Assessing normality with the help of Kolmogorov-Smirnov (K-S) test requires significant value higher than 0.05 for the data to be considered normally distributed. With the help of SPSS 26.0, the following Table 9 shows that Asymp. Sig. (2-tailed) appears to be 0.2. In other words, the data collected is normal.

Linearity Test
Assessing linearity of relationships between independent and dependent variables involved in this study is done by ensuring that the significant deviation from linearity is greater than 0.05. The following table show that Informativeness (INF) has a linear relationship with Attitude towards Advertisement (ATA):  (2021) Entertainment (ENT) and Attitude towards Advertisement (ATA) also appears to have linear relationship as shown in the table as follows:

Multicollinearity Test
Multicollinearity in this study is examined through the tolerance value and value of inflation factor (VIF). These values are shown in Table 11 below, demonstrating that there is no correlation between independent variables since the tolerance value exceeds 0.1 and VIF is kept under 10:

Heteroscedasticity Test
A good data set is one without heteroscedasticity which is characterized with significance value > 0.05. In table 14 below, it can be concluded that there is no heteroscedasticity detected in the data collected in this study:

Multiple Regression Analysis
Multiple regression analysis is carried out to find out how independent variables affect the dependent variable. The level of influence can be identified from the value of B that are listed in the following table: Given the values, the regression equation to describe the effect of all three independent variables to the dependent variable can be deduced as: There are several interpretations based on the identified regression equation, which are: a. It can be learnt that for each unit increase in Informativeness (X1), Attitude towards Advertisement (Y) will experience a rise as much as 0.348. b. Each unit increase of Entertainment (X2) results in 0.485 increase of Attitude towards Advertisement (Y). c. There will be a reduction as much as 0.259 each unit increase of Irritation (X3). d. Regardless, when neither of the independent variables change-it remains constant, Attitude towards Advertisement (Y) will still experience a constant increase as much as 5.115.
In conclusion, to generate better and more positive attitude among Generation Z towards their advertisement, Lokalate have to increase the informativeness and entertainment of their YouTube skippable ads while also decreasing its irritation.

Determination Test
With the help of SPSS 26.0 software, the extent to which an independent variable explains the dependent variable can be predicted. This is done by examining the value of R 2 . The R 2 of Informativeness (INF) is as shown: Irritation (IRR) appears to describe Attitude towards Advertisement (ATA) as much as 41.1%. This means that the other 58.9% is explained by other different factors. When all the three independent variables are combined, the R 2 value is:  (2021) The table above implies that Informativeness (INF), Entertainment (ENT) and Irritation (IRR) altogether has influences Attitude towards Advertisement (ATA) as much as 71.8%. The other 28.2% of the dependent variable, however, is described by other factors and variables which are not discussed in this study.

Hypothesis test: Partial t-test
To decide whether the proposed hypothesis should be accepted or rejected, there are two values that can be analyzed. First is the value of t counted while the other is value of significance, as compiled in the Table 20 below: Moreover, the value of significance is shown to be 0.000 which is less than 0.05. Hence, H1 is accepted which means that Informativeness (INF) of Lokalate YouTube skippable ad affects Generation Z attitude. b. Entertainment (ENT) has a tcount of 11.946 which exceeds ttable of 1.968. Value of significance is also observed to be 0.000 which is less than 0.05. H2 is therefore accepted, meaning that Entertainment (ENT) of Lokalate YouTube skippable ad affects Generation Z attitude. c. Irritation (IRR) has a tcount of -5.609 which is less than ttable -1.968.
Furthermore, the value of significance is 0.000, which is less than 0.05. Thus, H3 is accepted. This indicates that irritation (IRR) of Lokalate YouTube skippable ad affects Generation Z attitude.

DISCUSSION
Results of the tests demonstrates that there all hypothesis is accepted. Thus, there is a causality relationship between variables involved in this study; that is, Informativeness (INF), Entertainment (ENT) and Irritation (IRR) of Lokalate YouTube skippable ad partially affect the Generation Z Attitude towards Advertisement (ATA). This is finding supports the research conducted by Wiliyastuti et al., (2018), suggesting that informativeness and entertainment have a significant positive effect on millennial's attitude, while irritation affects negatively.
The effect of Informativeness (INF) on Attitude towards Advertisement (ATA) is observed to be positive, supporting the study results of Qousa & Wady (2018). In other words, an advertisement should be created as informative as possible to trigger a positive attitude from the audiences. Likewise, Entertainment (ENT) also has a positive effect on Attitude towards Advertisement (ATA). Consistent with the findings of Adidarma & Putri (2017), this suggests that positive attitude can be achieved by ensuring that entertaining elements are included in an advertisement. In contrast, Irritation (IRR) contributes to a negative effect on Attitude towards Advertisement (ATA), as deduced by Chungviwatanant et al. (2017) in their previous research. Hence, it is crucial for marketers to avoid advertisement that are irritating so that positive attitude from audiences can be established.

CONCLUSIONS AND RECOMMENDATIONS
All in all, marketers should pay attention on the informativeness, entertainment and irritation element their advertisement may possess. Reason is because all these properties will shape and determine how their audiences, Generation Z to be exact, will think, feel, and react towards the advertisement.
From on the elaborated discussion, several conclusions can be drawn: a. Informativeness, entertainment and irritation partially affect Generation Z attitude towards advertisement. b. Informativeness has a positive relationship with attitude towards advertisement, suggesting that higher level of informativeness in advertisement yields to more positive attitude. Inserting clear, concise, relevant as well as updated information are some ways to boost an advertisement's informativeness. c. Entertainment has a positive relationship with attitude towards advertisement, implying that more entertainment in advertisement generates more positive attitude. Some examples of entertaining elements that marketers may consider embedding in their advertisement include music, pictures, animations, or humour. d. Irritation has a negative relationship with attitude towards advertisement, meaning that higher degree of irritation will result in more negative attitude. This can be achieved by considering whether an advertisement is confusing, deceptive or appears to be disturbing and annoying to the audiences.

FURTHER STUDY
The determination tests shows that Informativeness (INF), Entertainment (ENT) and Irritation (IRR) does not completely explain Attitude towards Advertisement (ATA) as a dependent variable. Therefore, there are some other factors and variables that acts as an antecedent towards audience's attitude that is not included and discussed in this research. For future study, it is recommended to consider some other possible variables like personalization, or factors like income and education. Employing a more sophisticated data analysis technique such as Structural Equation Modelling (SEM) or data mining on this topic could also be carried out in the future to produce more comprehensive research finding.

ACKNOWLEDGEMENT
This research would not have been possible without the guidance of my advisor as well as my affiliated institution. I, therefore, would like to express my gratitude towards Mr. Genesis Sembiring and Universitas Pelita Harapan Medan Campus. For the opportunity to bring contribution through my research findings, I would also like to genuinely appreciate the journal editor and reviewer of Indonesian Journal of Business Analytics (IJBA).