which was in the tourism sector. In Indonesia, the tourism sector has a significant influence, as quoted from Kontan.co.id news entitled ‘PHRI: Kerugian Industri Pariwisata Karena Pandemi Sudah Lebih dari Rp 100 Trillium’ November 19, 2020, edition stated the tourism industry loss, hotels, and restaurants in Indonesia its value reached more than Rp100 trillion or US$7.1 billion as of early November 2020.

This loss in the tourism sector has had a big impact on Indonesia during the Covid-19 pandemic. Especially considering that Indonesia has many tourism attractions for domestic and foreign tourists. In the Kompas.com article entitled ‘5 Daerah Tujuan Wisata Favorit Saat Pandemi’ edition of 04 August 2020, there is a survey from GoLocal Domestic Travel for the digital travel platform Agoda. In the survey, five regions have become prima donna for domestic tourists after the Covid-19 pandemic, three of which are Bali, Bandung, and Yogyakarta.

Unfortunately, these three regions have become regions that have experienced the policy of the Implementation of Micro-Community Activity Restrictions (PPKM) based on March 8, 2021, Press Conference, which the Committee manages for Handling Covid-19 and National Economic Recovery. The decision to implement PPKM Micro in the area meets one of the four parameters as a limitation condition. These include: (1) The active case rate is above the national average; (2) The cure rate is below the national average, (3) The death rate is above the national average, and (4) The hospital occupancy rate for ICU and isolation rooms is above 70%.

In dealing with the crisis during the Covid-19 pandemic, the government initiated many movement efforts so that people comply with health protocols, one of which is the ‘Ingat Pesan Ibu' campaign. Reporting from Detik.com news entitled ‘Pemerintah Kampanyekan ‘Ingat Pesan Ibu’ Hadapi Covid -19, Ini Alasannya’ October 1, 2020, edition, the government raises the figure of mothers (Ibu) in campaigning for the dangers of spreading the Covid-19 disease, because compliance with this health protocol should be equated with public obedience to the mother's orders. It is done because of the diverse backgrounds of Indonesian people, but one thing in common is that everyone is born from a mother figure.

The content of the ‘Ingat Pesan Ibu’ campaign invites the Indonesian people to apply health protocol discipline through 3M activities. The 3M includes wearing a mask, maintaining a minimum distance of two meters, and washing hands with soap and running water.

Various media participated in realizing the ‘Ingat Pesan Ibu’ campaign in various forms, such as videos, posters, articles, podcasts, songs by Paid Reborn, and many more. The spread of this

campaign also occurred massively, both online through social media and mass media, and offline by displaying posters or installing audio in public places.

Charles U Larson (1992) in Venus (2019) explains three types of campaigns: product-oriented campaigns, candidate-oriented campaigns, and ideologically or cause-oriented campaigns. Product-oriented campaigns are usually motivated to gain financial and business benefits. Meanwhile, candidate-oriented campaigns are based on the desire to occupy a political position or gain the target audience's support. The ‘Ingat Pesan Ibu’ campaign is an ideologically or cause-oriented campaign that often has dimensions of social change or is often referred to as a social change campaign. This type of campaign aims to address social problems through the associated changes in public attitudes and behaviour. According to Hovland and Weis in Normawati et al. (2018), a change in a person's attitude has a more significant impact if caused by a communicator who has high credibility. In this case, the government, especially Covid-19 Task Force, the ‘Ingat Pesan Ibu’ campaign organizer, plays an essential role as a communicator.

Summarizing from several expert opinions, such as Bettinghaus (1973), Applbaum & Anatol (1976), Shimp & Delozier (1986), and Johnston (1994) in Normawati et al. (2018), mentioning aspects of the campaign consist of messages that can support the effectiveness of a message by emphasizes two critical aspects, namely message content and message structure. The message components that can be reviewed include material visualization, creativity in creating messages, emotional aspects, humour, and a direct approach to the target audience. In addition, it is worth looking at how the delivery of arguments can strengthen persuasion to the target audience in terms of the message structure. In the ‘Ingat Pesan Ibu campaign, the mother figure was chosen to ignite the emotional aspect so that the community can carry out health protocols such as obedience to the mother, mainly so that late adolescents can change behaviour more concretely and measurably.

In general, campaigns are persuasive activities that aim to influence patterns of thinking and behaving as expected. In research at the State Adolescent Health Resource Center at the University of Minnesota (2013), late adolescence begins to experience significant changes in thinking capacity. The transition from concrete to abstract thinking, late adolescence is increasingly able to understand and grapple with abstract ideas, think about possibilities, think ahead, think about thought processes, and put oneself in the shoes of others. In general, this changes their ability to think about themselves, others, and the world around them. The division of age groups or age categories issued by the Ministry of Health of the Republic of Indonesia (2009) in Amin and Dwi (2017) late adolescence is in the age range of 17-25 years.

The Central Statistics Agency (BPS) released the results of a survey on community behaviour during the implementation of the July 2021 edition of Emergency PPKM. The results stated that the behaviour of respondents in the Java-Bali region in implementing health protocols still needed attention. In the compliance index, avoiding crowds scored 22%, washing hands with soap or hand sanitizer was 25%, and maintaining a minimum distance of two meters as much as 33%. When compared to the level of compliance between the Java-Bali region and outside the Java-Bali region, respondents outside the Java-Bali region tend to be less compliant in implementing health protocols. The survey results stated that respondents who had not obeyed in avoiding crowds got a score of 31%, washing hands with soap or hand sanitizer as much as 35%, and maintaining a minimum distance of two meters as much as 44%.

Therefore, several factors influence the non-compliance in implementing health protocols. Afrianti and Rahmiati (2020) stated that compliance is influenced by age, education, knowledge, attitudes, and motivation. Adults are said to have a higher level of compliance than adolescents. This statement is supported based on the findings of data and facts in the field by the Covid-19 Task Force the field of Behaviour Change. There are data findings and facts in the field regarding implementing the July 2020 edition of the health protocol, which states that the obedience behaviour index shows that the higher the age, the more obedient the respondent behaves in fulfilling the appeal.

PT. Gojek Indonesia, an online transportation company, also carried out the implementation of a similar campaign. The campaign is #dirumahaja (#stayathome) to reduce the risk of transmission of Covid-19, which is spread through social media Instagram. Aprianita and Hidayat (2020) mention that in the implementation of the campaign, what attracts late adolescents when receiving campaign messages includes images, colours, and the novelty of content. Meanwhile, Naryoso et al. (2021) stated that digital campaigns must be packaged creatively as a strategic step to increase public participation.

In addition, Yuliarti (2021) stated the need to optimize campaigns both offline and online to prevent the spread of Covid-19. In offline campaigns, posters are better at conveying messages because they are dominated by images that make the creators express themselves, both in terms of content and artistically. In online campaigns, content dissemination through WhatsApp groups and Instagram is considered more effective in conveying campaign messages.

Based on this, a study is needed to determine the effectiveness of the ‘Ingat Pesan Ibu’ campaign as massive government moved to suppress the spread of Covid-19 in changing behaviour in late adolescent aged 17-25 years in the tourism areas of Bali, Bandung, and Yogyakarta.

This paper hopes that an effective campaign will be realized to achieve behaviour change under government recommendations, significantly to reduce the risk of Covid-19 among Indonesian late adolescent, especially in tourist areas affected by the pandemic Bali, Bandung, and Yogyakarta. In addition, it is also hoped that there will be a novelty in the preparation of the structure and content of a compelling message to create a creative and on-target campaign. The scientific benefit of this paper is that it can be used as a material or reference for further research.

The purpose of the study was to determine whether the ‘Ingat Pesan Ibu’ campaign was influential in changing late adolescent behaviour. Furthermore, to find out how far the ‘Ingat Pesan Ibu’ campaign was received by the late adolescent in the tourism areas of Bali, Bandung, and Yogyakarta, which are three of the five regions favourite tours affected by the pandemic.

This study uses quantitative research methods. According to Sugiyono (2018) quantitative research methods are defined as research methods based on the philosophy of positivism used to examine specific populations or samples. Collecting data using research instruments, data analysis is quantitative/statistical, intending to describe and test established hypotheses. The test step was carried out to measure changes in the behaviour of late-level adolescents aged 17-25 years as a single variable X using the Ostergaard campaign model.

 

2. MATERIALS AND METHODS

2.1. CAMPAIGN MODEL

This study uses a campaign model developed by Leon Ostergaard. According to Ostergaard in Venus (2019), a campaign program design must be supported by scientific findings to overcome the social problems faced. Therefore, the campaign program should always start from problem identification or what is also known as the pre-campaign stage.

The first step taken by campaign makers is to identify the perceived real problems, for example, the high number of Covid-19 cases in Indonesia. From this example, a cause-and-effect relationship is sought with the existing facts. The second stage is campaign management, starting from design, implementation, to evaluation.

Research is carried out to formulate messages, campaign actors, channels, and appropriate campaign implementation techniques in the identification process. It is not enough; campaign makers also need to research the target audience in order to be able to formulate a compelling message. Venus (2019) argues that campaign messages have various dimensions, including verbal, non-verbal, and visual messages. In general, whatever the dimensions, message construction must be based on considerations of simplicity, familiarity with the audience's situation, clarity, conciseness, novelty, consistency, courtesy, and conformity with the campaign object. Simplicity can make the message easy to understand and remember (memorability).

At the management stage, the entire content of the campaign program is directed to equip and influence aspects of the knowledge, attitudes, and skills of the target audience. These three aspects in the scientific literature are prerequisites for behaviour change. The knowledge aspect is focused on growing awareness of the target audience, which refers to the next stage, namely attitude change. At this advanced stage, the target audience is expected to sympathize with the issues raised through the campaign. If this has been fulfilled, a measurable review of the actions or skills of the target audience will be carried out.

The last stage of Ostergaard's model is the evaluation stage of reduced problems, called the post-campaign stage. It will be evaluated whether the message reaches the audience? Can they remember the message? Or can they receive the message? Evaluation is directed at the campaign's effectiveness in resolving or reducing the problems as identified in the pre-campaign stage.

The following is a model of the Ostergaard campaign presented in the chart below.

Figure 1 Ostergaard Campaign Model Chart Source: Venus (2019)

 

Klingemann and Rommele (2002) in Venus (2019) define campaign channels as all forms of the medium used to convey messages to audiences. The form can be paper to write messages, telephone, public dialogue, counselling, posters, banners, internet/social media, radio, television, or newspapers. In communication campaigns, mass media is used as the primary communication channel because it has many audiences. According to Rogers (1987) in Venus (2019) the role of mass media in campaigns is essential because of the campaign's target in the crowd, the public, and the community. The mass media, which are often used as the primary communication channel in communication campaigns, can multiply the dissemination of information and persuade the audience. Regarding the ability of mass media to influence attitudes, opinions, and behaviour of audiences, Clapper in Venus  (2019) distinguishes six types of changes that may arise due to the use of mass media, namely: (1) causing the desired change (conversion); (2) cause unwanted changes; (3) cause small changes (both in form and intensity); (4) smoothing change (desired or not); (5) reinforce what is (no change); and (6) prevent change.

Rice and Atkin (2009) in Venus (2019) state that campaigns have begun to migrate from mass media to social media, which are considered more interactive and build engagement among users. This prediction is supported by the rapid penetration of the internet, which has become a vehicle for social media in various parts of the world. From the World Stats Internet site, Indonesia is ranked 4th with the most significant number of internet users in the world from 2020 to the first   quarter of 2021 Internetworldstats.com. (2021). Based on this number, it is undeniable that social media has become popular as a campaign channel.

Venus (2019) ensures that social media will determine future campaigns, seeing many social media users and the increasing trend in Indonesia. Social media has advantaged that mass media does not have, such as providing space for interaction, eliciting participation, involvement for its users, and even individual intervention at the grassroots in a campaign message.

 

2.2. EFFECTIVE CAMPAIGN

According to Suprapto (2009), campaign communication observes what the actual problems that are developing in the community are, and the aim is to provide solutions to public problems while at the same time explaining the steps and ways to solve public problems that are being faced.

The campaign's effectiveness can be seen from the fulfillment of the objectives achieved in the campaign process. According to Normawati et al. (2018), the campaign is broadly a persuasive activity to influence the expected mindset, attitudes, and behaviour. Campaigns in communication inform, persuade, and motivate changes in audience behaviour. In addition, Rochimah (2009) states that behaviour change will occur if the perpetrator benefits when changing his old behaviour. Behaviour change is a process, not only a change in the level of behaviour but also the level of knowledge or understanding. Thus, the factors and processes of change help us understand the various levels of behaviour change that are quite diverse and not only seen from the final result, namely action. For instance, from the emergence of awareness, the growth of interest, and the emergence of attraction to the communicant who indeed responds to a message differently.

To determine the effectiveness of achieving campaign objectives, evaluations are carried out, which often focus on campaign effectiveness. According to Venus (2019) the evaluation is carried out not only on objectives based on the effects of the tested campaigns but also effects-free campaign objectives. In evaluating the effects of a campaign, campaigners must consider the quality and quantity of the efforts and activities carried out during the campaign, namely what the campaigners have done and how well they did it. Campaigners must also see the ongoing process of the campaign in testing the effects, namely what is done and what is not done.

The sampling technique used is cluster sampling, carried out in 2 (two) stages, namely the first stage: determining the sample area (Bali Province, Bandung City, and Yogyakarta City). Followed by the second stage: determining the research population taken, namely late adolescents aged 17-25 years. From this population, a sample was drawn using the Taro Yamane formula Sugiyono (2018) as many as 100 respondents per region with an error tolerance of 10%. The data collection method used is an online questionnaire through the Google Form platform and distributed on 11-14 February 2021. This questionnaire contains a Likert measurement scale based on Leon Ostergaard's campaign model.

The data analysis technique used is descriptive statistics, including the table of the mean (mean), standard deviation (standard deviation), which is processed using the SPSS version 25 program, bar chart, and the trend of the data towards the data resulting from research activities which are then analysed descriptively. Referring to Sugiyono (2018) the statistical technique for testing descriptive hypotheses is the One-Sample T-Test using interval data to see the trend of the data. In this study, there is a basis for making decisions on the One-Sample T-Test: If the value of Sig. (2-tailed) < 0.1, then H₀ is rejected. But if the value of Sig. (2-tailed) > 0.1, then H0 is accepted.

Sugiyono (2018) states that a valid and reliable instrument is an absolute requirement to obtain valid and reliable research results. The validity test uses the Pearson Product Moment formula to show the level of instrument validity. The basis for decision making in the Pearson Product Moment validity test is to compare R-value and r-table with the decision: If the value of R-value > r-table, then the item is declared valid. However, if the value of R-value < r-table, the question item is declared invalid.

Furthermore, the reliability test uses the Cronbach Alpha formula to support the formation of validity. The instrument reliability criteria: r₁ > 0.90 can be said to be perfect reliability; 0.90 > r₁ > 0.70 can be said to be high reliability; 0.70 > r₁ > 0.50 can be said to be moderate reliability; and r₁ > 0.50 can be said to be low reliability.

Furthermore, the following chart is displayed to show the flow of the frame of mind regarding the effectiveness of the ‘Ingat Pesan Ibu’ campaign.

Figure 2 Research Framework Source: Proceed by the Researchers

 

3. RESULTS AND DISCUSSIONS

From the questionnaires that have been distributed, there are 300 respondents with an equally distributed distribution, namely 100 respondents in each area of Bali, Bandung, and Yogyakarta. Based on the data that has been collected, it is found that women in each region dominate the respondents. A total of 79 people from the Yogyakarta area, 59 people in the Bandung area, and 67 people from the Bali area. At the same time, male respondents were 21 people in the Yogyakarta area, 41 people in the Bandung area, and 33 people in the Bali area.

Most respondents are between 20-22 years in each region, with the highest number in the Bali area as many as 80 people, followed by the Bandung area as many as 77 people, and the Yogyakarta area as many as 54 people. For educational background data, most respondents are currently pursuing undergraduate education in each region, namely 53 people in the city of Yogyakarta, 85 people in the city of Bandung, and 70 people in the province of Bali.

The questionnaire distribution contains a list of statements that include a single variable X to measure the behaviour change of late adolescent (17-25 years) towards the message of the 'Ingat Pesan Ibu' campaign. The indicator variable X includes knowledge (X₁), attitudes (X₂), skills (X₃), and behaviour (X₄) according to the Ostergaard campaign model. There is a measurement scale used in answering the list of questionnaire statements, namely a scale of 1-5 with the following information: 1 = Strongly Disagree, 2 = Disagree, 3 = Neutral, 4 = Agree, and 5 = Strongly Agree.

The indicators include knowledge (X₁) regarding understanding the importance of 3M (X_1.1), knowing that the 'Ingat Pesan Ibu' campaign was initiated by the Covid-19 Task Force (X_1.2), increasing awareness of the importance of health protocols (X_1.3), and know good and correct health protocols (X_1.4).

Indicators that include attitudes (X₂) consist of respondents' acceptance of the ‘Ingat Pesan Ibu' campaign message (X_2.1), having an upbeat assessment of the ‘Ingat Pesan Ibu' campaign (X_2.2), and complying with the health protocol after reading the ‘Ingat Pesan Ibu' campaign. (X_2.3).

Then the indicator, which includes skills (X₃) regarding the ‘Ingat Pesan Ibu' campaign, encourages respondents to communicate messages about the importance of health protocols to their surroundings (X_3.1), respondents are more careful in implementing health protocols (X_3.2) and foster a sense of responsibility on the safety of people around the informant (X_3.3).

The indicators include behaviour (X₄) regarding the application of health protocols in daily life (X_4.1), the success of consistently implementing health protocols (X_4.2), the ‘Ingat Pesan Ibu' campaign helping respondents understand the situation and conditions of the Covid-19 pandemic (X_4.3), respondents have taken off masks in public rooms (X_4.4), respondents always wash their hands during activities (X_4.5), respondents always carry hand sanitizer (X_4.6), respondents always carry antiseptic wet wipes (X_4.7), respondents always keep their distance in public places (X_4.8), and respondents adopt a healthy lifestyle (X_4.9).

In showing the level of validity of the instrument, the validity test is carried out in each region first. Based on the results of the calculation of the validity of each statement item, all statement items are declared valid because r_x > r-table or the value of r_x is greater than the value of r-table, with the smallest r_x in each region being in the variable X_4.4.

Table 1 Table of Validity Test Results for Bali, Bandung, and Yogyakarta
Bali Validity Test Results
X	rx	r-table	Description
X1.1	0.807	0,256	VALID
X1.2	0.596	0,256	VALID
X1.3	0.781	0,256	VALID
X1.4	0.848	0,256	VALID
X2.1	0.74	0,256	VALID
X2.2	0.851	0,256	VALID
X2.3	0.845	0,256	VALID
X3.1	0.871	0,256	VALID
X3.2	0.896	0,256	VALID
X3.3	0.862	0,256	VALID
X4.1	0.746	0,256	VALID
X4.2	0.81	0,256	VALID
X4.3	0.766	0,256	VALID
X4.4	0.32	0,256	VALID
X4.5	0.822	0,256	VALID
X4.6	0.78	0,256	VALID
X4.7	0.767	0,256	VALID
X4.8	0.793	0,256	VALID
X4.9	0.77	0,256	VALID
Bandung Validity Test Results
X	rx	r-table	Description
X1.1	0.829	0,256	VALID
X1.2	0.763	0,256	VALID
X1.3	0.811	0,256	VALID
X1.4	0.763	0,256	VALID
X2.1	0.801	0,256	VALID
X2.2	0.798	0,256	VALID
X2.3	0.804	0,256	VALID
X3.1	0.837	0,256	VALID
X3.2	0.912	0,256	VALID
X3.3	0.879	0,256	VALID
X4.1	0.83	0,256	VALID
X4.2	0.857	0,256	VALID
X4.3	0.818	0,256	VALID
X4.4	0.282	0,256	VALID
X4.5	0.826	0,256	VALID
X4.6	0.85	0,256	VALID
X4.7	0.794	0,256	VALID
X4.8	0.861	0,256	VALID
X4.9	0.737	0,256	VALID
Yogyakarta Validity Test Results
X	rx	r-table	Description
X1.1	0.823	0,256	VALID
X1.2	0.764	0,256	VALID
X1.3	0.836	0,256	VALID
X1.4	0.814	0,256	VALID
X2.1	0.848	0,256	VALID
X2.2	0.817	0,256	VALID
X2.3	0.773	0,256	VALID
X3.1	0.855	0,256	VALID
X3.2	0.86	0,256	VALID
X3.3	0.873	0,256	VALID
X4.1	0.772	0,256	VALID
X4.2	0.725	0,256	VALID
X4.3	0.738	0,256	VALID
X4.4	0.339	0,256	VALID
X4.5	0.656	0,256	VALID
X4.6	0.732	0,256	VALID
X4.7	0.78	0,256	VALID
X4.8	0.75	0,256	VALID
X4.9	0.772	0,256	VALID

 

After the validity test is done, the reliability test is carried out. A reliability test is needed to support the formation of validity. The calculation of the reliability test is the same as the validity test, which is also carried out in each region of Bali, Bandung, and Yogyakarta.

Based on the results of the calculation of the reliability test, it can be categorized as perfect reliability because the r1 value> 0.90 so that the instrument is feasible as a measuring tool in measuring the effectiveness of the 'Ingat Pesan Ibu' Campaign in Behaviour Change for Late Adolescents (17-25 years) in the Bali Tourism Area, Bandung, and Yogyakarta.

Table 2 Table of Reliability Test Results for Bali, Bandung, and Yogyakarta
Region	Cronbach’s Alpha	Description
Bali	0.941	Perfect Reliability
Bandung	0.926	Perfect Reliability
Yogyakarta	0.919	Perfect Reliability

 

Based on data, the number of respondents who have encountered the 'Ingat Pesan Ibu' campaign on various campaign channels, the majority found it on social media and TV programs. In the Yogyakarta area, 41 people met on social media and 36 people on TV programs. For the Bandung area, 37 people found it on social media and 32 people on TV programs. As for the Bali area, 39 people found it on social media, and 34 people found it on TV programs.

As for the respondents in the Bali area, many of the 'Ingat Pesan Ibu' campaigns were found through music, with as many as 77 people. Meanwhile, social media Facebook and Tiktok were the channels that the least number of respondents found in displaying the 'Ingat Pesan Ibu' campaign, as many as 81 respondents.

The results in the Bandung area show similarities to the Bali area, respondents in the Bandung area often encounter the 'Ingat Pesan Ibu' campaign through music as many as 77 people. Meanwhile, social media Facebook is the channel with the least number of respondents, namely 86 people.

For the Yogyakarta area, it was found that many respondents encountered the ‘Ingat Pesan Ibu’ campaign through Instagram social media as many as 74 people and were followed by TV programs as many as 73 people. Meanwhile, the channel that most respondents found in presenting the ‘Ingat Pesan Ibu’ campaign was the same as in the Bandung area, namely social media Facebook as many as 70 respondents.

After the respondent's initial data has been collected, in the effectiveness analysis section using the pretest-posttest scores, which are quantitative data, it is necessary to test the normality test and One-Sample T-test.

Normality test aims to test the data whether the data is normally distributed or not. The normality test was carried out using the Kolmogorov-Smirnov normality test in this study. As for the normality test results, it is known that the data is typically distributed. Then the T-test is carried out to test the existing hypotheses. The following table shows the results of the Kolmogorov-Smirnov normality test.

Table 3 Table of Normality Test Results
One-Sample Kolmogorov-Smirnov Test
		Bali	Bandung	Yogyakarta
N		100	100	100
Normal Parametersa,b	Mean	63.8949	63.536	64.3308
	Std. Deviation	12.30289	11.94935	11.47984
Most Extreme Differences	Absolute	0.107	0.075	0.075
	Positive	0.075	0.074	0.067
	Negative	-0.107	-0.075	-0.075
Test Statistic		0.107	0.075	0.075
Asymp. Sig. (2-tailed)		.007c	.175c	.185c

                                       

H₀ = The ‘Ingat Pesan Ibu’ campaign effectively changes the behaviour of late adolescents in the tourism areas of Bali, Bandung, and Yogyakarta.

H₁ = The ‘Ingat Pesan Ibu’ campaign is not effective in changing the behaviour of late adolescents in the tourism areas of Bali, Bandung, and Yogyakarta.

To see whether the ‘Ingat Pesan Ibu’ campaign effectively changes the behaviour of late adolescents in the tourism areas of Bali, Bandung, and Yogyakarta, a T-test calculation was carried out using one-sample test in each region.

Based on the table, T-test results in the Bali region shows the value of Sig. (2-tailed) is greater than 0.1 (0.997 > 0.1) which means H0 is accepted. For the Bandung area, the T-test results show the value of Sig. (2-tailed) is greater than 0.1 (0.976 > 0.1) which means H0 is accepted. The T-test results for the Yogyakarta region also show similarities with the previous regions, that the value of Sig. (2-tailed) is greater than 0.1 (0.999 > 0.1) which means H0 is accepted. It means that the T-test results of the three regions show that H0 is equally accepted and H1 is rejected.

Table 4 T-Test Results for the Regions of Bali, Bandung, and Yogyakarta
One Sample Test
Region	t	df	Sig.(2-tailed)	Mean Difference	90% Confidence Interval of the Difference
					Lower	Upper
Bali	0.004	99	0.997	0.00491	-2.0379	2.0477
Bandung	0.03	99	0.976	0.036	-1.9481	2.0201
Yoyakarta	0.001	99	0.999	0.00082	-1.9053	1.9069

 

After carrying out the submission step and carrying out descriptive statistics, including the mean table and standard deviation was processed using SPSS version 25 to see the trend of the data towards the data resulting from research activities to be analysed descriptively.

Bali's descriptive statistical table shows that knowledge (X₁), attitudes (X₂), skills (X₃) have a mean above number 4, which indicates that the ‘Ingat Pesan Ibu’ campaign is effective in changing the behaviour of respondents up to 3 (three) stages, namely knowledge (X₁), attitudes (X₂), and 

skills (X₃). The X_4.4 variable evidence this with a mean value of 2.92 approaching the number 3, which shows that respondents are hesitant to take off masks in public rooms.

The most significant standard deviation value is 1,277, found in the X4.4 variable, with the indicator that the respondent has taken off a mask in a public room. Namely, as many as 17 respondents answered Strongly Disagree, 21 respondents answered Disagree, 28 respondents answered Neutral, 21 respondents answered Agree, and 13 respondents answered Strongly Agree. From this data, it is found that the X4.4 variable is the most heterogeneous data.

The minor standard deviation of 0.658 is found in the X2.1 variable with the indicator of respondents' acceptance of the campaign message ‘Ingat Pesan Ibu’. A total of 9 respondents answered Neutral, 35 respondents answered Agree, and 56 respondents answered Strongly Agree. From this data, it is found that the X2.1 variable is the most homogeneous data.

Table 5 Descriptive Statistics of Bali Province
Descriptive Statistics
N		Minimum	Maximum	Mean	Std. Deviation
X1.1	100	1	5	4.29	0.769
X1.2	100	1	5	4.02	1.025
X1.3	100	1	5	4.16	0.896
X1.4	100	1	5	4.27	0.827
X2.1	100	3	5	4.47	0.658
X2.2	100	2	5	4.52	0.703
X2.3	100	1	5	4.32	0.931
X3.1	100	1	5	4.08	0.918
X3.2	100	2	5	4.16	0.825
X3.3	100	1	5	4.25	0.796
X4.1	100	1	5	4.27	0.886
X4.2	100	1	5	4.11	1.004
X4.3	100	1	5	4.16	0.972
X4.4	100	1	5	2.92	1.277
X4.5	100	1	5	4.2	0.865
X4.6	100	1	5	4.29	1.008
X4.7	100	1	5	3.62	1.144
X4.8	100	1	5	4.11	0.994
X4.9	100	1	5	4.22	0.883
Valid N (listwise)	100

                        

The results in the city of Bandung show similarities to the province of Bali. It is because knowledge (X₁), attitudes (X₂), skills (X₃) have a mean value above 3.5 which indicates that the ‘Ingat Pesan Ibu’ campaign is effective in changing the behaviour of respondents up to three stages, namely knowledge (X₁), attitudes (X₂), and skills (X₃). To arrive at the behaviour stage (X₄), it is considered still lacking because there is a mean value still at number 3, namely the X_4.4 variable with a mean value of 3.29, indicating that respondents are hesitant ever to take off masks in public rooms.

Variable X_4.4 is also the most significant standard deviation at 1.343 with the indicator that the respondent has taken off a mask in a public room, as many as 12 respondents answered Strongly Disagree, 21 respondents answered Disagree, 15 respondents answered Neutral, 30 respondents answered Agree, and 22 respondents answered Strongly Agree. From this data, it is found that the X4.4 variable is the most heterogeneous data.

While the minor standard deviation of 0.611 is found in the X_2.2 variable with a positive assessment indicator of the homogeneous ‘Ingat Pesan Ibu’ campaign. A total of 6 respondents answered Neutral, 39 respondents answered Agree, and 55 respondents answered Strongly Agree.

Table 6 Descriptive Statistics of Bandung City
Descriptive Statistics
N		Minimum	Maximum	Mean	Std. Deviation
X1.1	100	3	5	4.34	0.67
X1.2	100	2	5	4.15	0.833
X1.3	100	2	5	4.27	0.723
X1.4	100	2	5	4.27	0.723
X2.1	100	2	5	4.35	0.687
X2.2	100	3	5	4.49	0.611
X2.3	100	2	5	4.18	0.77
X3.1	100	2	5	3.94	0.874
X3.2	100	2	5	4.2	0.696
X3.3	100	2	5	4.26	0.76
X4.1	100	2	5	4.21	0.756
X4.2	100	1	5	3.99	0.882
X4.3	100	2	5	4.25	0.783
X4.4	100	1	5	3.29	1.343
X4.5	100	1	5	4.15	0.869
X4.6	100	1	5	4.18	0.857
X4.7	100	1	5	3.71	1.166
X4.8	100	1	5	4.14	0.853
X4.9	100	1	5	4.03	0.937
Valid N (listwise)	100

                                                      

The city of Yogyakarta also shows similarities with the two previous regions. The variables knowledge (X₁), attitudes (X₂), skills (X₃) have a mean value above number 4, which indicates the 'Ingat Pesan Ibu' campaign is effective in changing the behaviour of respondents up to three stages, namely knowledge (X₁), attitudes (X₂), and skills (X₃). To reach at the behaviour stage (X₄), it is considered lacking because there is a mean value still at number 3, namely the X_4.4 variable with a mean value of 3.12, indicating that respondents are hesitant to take off masks in public rooms.

Similar to the previous results, the variable X_4.4 is also the most significant standard deviation. In the Yogyakarta area, the results showed 1,402 for the X_4.4 variable, indicating that the respondent had taken off a mask in a public room. Namely, as many as 16 respondents answered Strongly Disagree, 20 respondents answered Disagree, 24 respondents answered Neutral, 16 respondents answered Agree, and 24 respondents answered Strongly Agree. From this data, it is found that the X_4.4 variable is the most heterogeneous data.

While the minor standard deviation of 0.643 is found in the X_2.2 variable with a positive assessment indicator of the ‘Ingat Pesan Ibu’ campaign. A total of 1 respondent answered Strongly Disagree, 5 respondents answered Neutral, 40 respondents answered Agree, and 54 respondents answered Strongly Agree. From this data, it is found that the X_2.2 variable is the most homogeneous data.

Table 7 Descriptive Statistics of Yogyakarta City
Descriptive Statistics
N		Minimum	Maximum	Mean	Std. Deviation
X1.1	100	1	5	4.4	0.829
X1.2	100	2	5	4.06	0.886
X1.3	100	1	5	4.29	0.935
X1.4	100	1	5	4.32	0.827
X2.1	100	2	5	4.3	0.81
X2.2	100	2	5	4.47	0.643
X2.3	100	1	5	4.17	0.842
X3.1	100	1	5	4.1	0.937
X3.2	100	1	5	4.19	0.775
X3.3	100	2	5	4.28	0.805
X4.1	100	1	5	4.32	0.777
X4.2	100	2	5	4.19	0.8
X4.3	100	1	5	4.23	0.874
X4.4	100	1	5	3.12	1.402
X4.5	100	1	5	4.16	0.907
X4.6	100	1	5	4.27	0.962
X4.7	100	1	5	3.74	1.276
X4.8	100	2	5	4.2	0.865
X4.9	100	1	5	4.09	0.944
Valid N (listwise)	100

                                                      

4. CONCLUSIONS AND RECOMMENDATIONS

From the research results in each region, it can be seen that the T-test (T-test) shows the value of Sig. (2-tailed) is more significant than 0.1 and accepts H₀, which means that the ‘Ingat Pesan Ibu’ campaign effectively changes the behaviour of late adolescents (17-25 years) in the tourism area of Bali, Bandung, Yogyakarta. Meanwhile, the results of descriptive statistics, which include a table of the mean and standard deviation of each region, show that the ‘Ingat Pesan Ibu’ Campaign was delivered well and changed the behaviour of late adolescents to 3 (three) stages, namely knowledge (X₁), attitudes (X₂), and skills (X₃). To reach at the behaviour stage (X₄), it is stated that it cannot be perfect because the mean value result is still at number 3 meaning the respondent still have doubt in implementing the health recommendation as stated by ‘Ingat Pesan Ibu’ campaign. It is also supported by the X_4.4 variable, which shows the most heterogeneous data to get various answers about removing masks in public spaces.

This research can provide an overview of the impact of implementing the ‘Ingat Pesan Ibu’ campaign in the tourism areas of Bali, Bandung, and Yogyakarta. In this case, it mainly describes the extent to which the implementation of the ‘Ingat Pesan Ibu’ campaign is effective among late adolescents in the research area so that the results obtained can be used as evaluation material for campaign organizers, especially the Covid-19 Task Force continuing or designing health campaigns in the future.

This research is expected to help design an effective campaign. However, further study is needed. Like research on which media is the most effective campaign channel, considering that this research only discusses various forms of campaign channels. There is a need for detailed research on each campaign channel specifically, primarily through the song ‘Ingat Pesan Ibu’ by Padi Reborn. In addition, further research can also raise the comparison of acceptance of campaign messages in terms of gender because women generally dominate this research. Furthermore, research can also be completed by analysing other areas affected by the pandemic to find a more comprehensive view of this campaign.

 

ACKNOWLEDGEMENTS

The authors would like to thank Directorate of Learning and Student Affairs, Directorate General of Higher Education for funding this scientific article in 2021. The authors also would like to thank Telkom University, the Faculty of Communication and Business, and the Communication Studies Study Program for providing the opportunity to incorporate this scientific article into the International Journal of Research Granthaalayah with Dr. Roro Retno Wulan, S.Sos., M.Pd, as well as family and PKM-AI colleagues who have encouraged realizing this scientific article.

 

REFERENCES

Afrianti, N., & Rahmiati, C. (2020). Faktor-Faktor yang Mempengaruhi Kepatuhan Masyarakat terhadap Protokol Kesehatan Covid-19. [Accessed on August 13, 2021] Retrieved from journal.stikeskendal.ac.id/index.php/PSKM/article/view/1045

Al Amin, Muhammad & Dwi Juniati. (2017). Klasifikasi Kelompok Umur Manusia Berdasarkan Analisis Dimensi Fraktal Box Counting dari Citra Wajah dengan Deteksi Tepi Canny. Retrieved from jurnalmahasiswa.unesa.ac.id/index.php/mathunesa/article/view/19398/17715. [Accessed on January 07, 2021] Retrieved from https://ejournal.unesa.ac.id/index.php/mathunesa/article/view/19398

Aprianita & Hidayat. (2020). Analisis Pesan Kampanye #dirumahaja di Tengah Pandemi Covid-19. Retrieved from jurnal.uinsu.ac.id/index.php/KOMUNIKOLOGI/article/view/7910. [Accessed on August 13, 2021] Retrieved from https://doi.org/10.30829/komunikologi.v4i2.7910

Badan Pusat Statistik. (2021). Perilaku Masyarakat Pada Masa PPKM Darurat, Hasil Survei Perilaku Masyarakat pada Masa Pandemi Covid-19, Periode 13-20 Juli 2021 [Set data]. [Accessed on August 13, 2021] Retrieved from www.bps.go.id/publication/2021/08/02/29234b08faa4910dee5279af/perilaku-masyarakat-pada-masa-ppkm-darurat--hasil-survei-perilaku-masyarakat--pada-masa-pandemi-covid-19--periode-13-20-juli-2021.html

Detik.com (2020). Pemerintah Kampanyekan 'Ingat Pesan Ibu' Hadapi Covid-19, Ini Alasannya. [Accessed on February 04, 2021] Retrieved from https://news.detik.com/berita/d-5196248/pemerintah-kampanyekan-ingat-pesan-ibu-hadapi-covid-19-ini-alasannya

Detik.com (2021). Alert ! 3 daris 4 Varian Corona 'Ganas' Sudah di Indonesia. [Accessed on August 12, 2021] Retrieved from https://health.detik.com/berita-detikhealth/d-5567273/alert-3-dari-4-varian-corona-ganas-sudah-ada-di-indonesia

Internetworldstats.com. (2021). Top 20 Countries With The Highest Number of Internet User. [Accessed on August 13, 2021] Retrieved from : www.internetworldstats.com/top20.html   

Kompas.com (2020). 5 Daerah Tujuan Wisata Favorit Wisatawan Saat Pandemi, Bali Teratas. [Accessed on February 04, 2021] Retrieved from https://travel.kompas.com/read/2020/08/04/161000127/5-daerah-tujuan-wisata-favorit-wisatawan-saat-pandemi-bali-teratas?page=all   

Kontan.co.id (2020). PHRI : Kerugian Industri Pariwisata Karena Pandemi Sudah Lebih Dari Rp 100 Triliun. [Accessed on August 12, 2021] Retrieved from industri.kontan.co.id/news/phri-kerugian-industri-pariwisata-karena-pandemi-sudah-lebih-dari-rp-100-triliun

Naryoso, A., Febriyani, A., & Kaloka, R. (2021). Digital Campaign to Reduce Covid-19 Pandemic Risk. Retrieved from https://journal.umy.ac.id/index.php/jkm/article/view/9811. [Accessed on August 13, 2021] Retrieved from https://doi.org/10.18196/jkm.131046

Normawati, et al. (2018). Pengaruh Kampanye "Let's Disconnect To Connect" Terhadap Sikap Anti Phubbing (Survei pada Followers Official Account Line Starbucks Indonesia). [Accessed on January 07, 2021] Retrieved from researchgate.net/publication/329536120_INFLUENCE_OF_THE_CAMPAIGN_LET'S_DISCONNECT_TO_CONNECT_ON_ANTI-PHUBBING_ATTITUDE_SURVEY_IN_LINE_STARBUCKS_INDONESIA_OFFICIAL_ACCOUNT_FOLLOWERS

Rochimah. (2009). Evaluasi Pelaksanaan Kampanye Sosial Perilaku Hidup Bersih dan Sehat untuk Menurunkan Angka Diare di Kabupaten Kulonprogo. Retrieved from https://media.neliti.com/media/publications/101082-ID-evaluasi-pelaksanaan-kampanye-sosial-per .pdf. [Accessed on January 28, 2021] Retrieved from https://ojs.uajy.ac.id/index.php/jik/article/view/207

Satuan Tugas Penanganan Covid-19 Bidang Perubahan Perilaku. (2020). Strategi dan Konsep Dasar Perubahan Perilaku. [Accessed on February 14, 2021] Retrieved from bersamahadapikorona.kemdikbud.go.id/wp-content/uploads/2020/09/Konsep-Dasar-Perubahan-Perilaku-30-Agustus-2020-v6-1.pdf

State Adolescent Health Resource Center. (2013). Understanding Adolescent Seeing Through A Development Lens. [Accessed on January 07, 2021] Retrieved from amchp.org/programsandtopics/AdolescentHealth/projects/Documents/SAHRC%20AYADevelopment%20LateAdolescentYoungAdulthood.pdf

Sugiyono. (2018). Metode Penelitian Kuantitatif. Bandung : Alfabeta.

Suprapto. (2009). Pengantar Teori dan Managemen Komunikasi. [Accessed on February 04, 2021] Retrieved from https://books.google.co.id/books?hl=en&lr=&id=xtHs4pLWdqAC&oi=fnd&pg=PA1&dq=teori+kampanye&ots=JhPYCGYKfa&sig=Z1VGEGJv4a8V_pUqnFajxf2HqAs&redir_esc=y#v=onepage&q=kampanye&f=false

Venus, Antar. (2019). Managemen Kampanye : Panduan Teoritis dan Praktis dalam Mengefektifkan Kampanye Komunikasi Publik. Bandung : Simbiosa Rekatama Media. 

World Health Organization. (2020). Coronavirus disease (COVID-19) Question and Answers. [Accessed on January 07, 2021] Retrieved from who.int/emergencies/diseases/novel-coronavirus-2019/question-and-answers-hub/q-adetail/coronavirus-disease-covid-19 

Yuliarti, M. S. (2021). Optimalisasi Kegiatan Online dan Offline dalam Kampanye Pencegahan COVID-19. Retrieved from ojs.unpkediri.ac.id/index.php/PPM/article/view/14912/2106 [Accessed on August 13, 2021] Retrieved from https://doi.org/10.29407/ja.v5i1.14912