Project topics and materials
Get Complete Project Material File(s) Now! »
Methodology
The methodology outlines the process of data collection for approaching the research questions. First, the research philosophy and approach are presented. This is followed by the strategies, choices, and time horizon selected. Besides, the techniques and procedures, explaining the survey design, target group, sampling process, and data analysis design are indicated. Finally, the ethicality, reliability, and validity of the study are highlighted. This section shows how the research was undertaken by describing the theoretical and philosophical assumptions as well as the methods adopted.
Research philosophy
Neville (2007) states that research is not neutral, because if reflects a mixture of the researchers’ interests, assumptions, values, abilities, aims, and ambitions. Therefore, embracing a research philosophy is necessary. A research philosophy illustrates the set of beliefs regarding the reality of the investigated nature (Bryman, 2012). The research philosophy reveals essential assumptions concerning the researchers’ perspective of the world. Accordingly, the research strategy and methods are based on these assumptions. The adopted philosophy could be influenced by practical considerations. The core influence of the adopted philosophy is the perspective of the researchers on the connection between the developed process and the knowledge (Saunders, Lewis, & Thornhill, 2009).
The philosophy used in this research was realism. Realism relates to scientific inquest. “The essence of realism is that what the senses show us as reality is the truth: that objects have an existence independent of the human mind” (Saunders et al., 2009, p. 114). The focus lies upon explaining within a context or contexts. The phenomenon of nostalgic marketing is a broad context. This research examined if millennials are nostalgic and if the use of nostalgic advertising, a nostalgic marketing approach, positively influences brand attitude and purchase intention. The effects of nostalgia on brand attitude and purchase intention among millennials is, therefore, explained within the context of nostalgic marketing. Moreover, with realism as the research philosophy, the researchers are biased by world views, cultural experiences, and upbringing (Saunders et al., 2009). The choice of realism philosophy is further fortified by the fact that both researchers are part of the millennial generation themselves.
Research approach
This research made use of a positivistic approach. A positivistic approach is based on research methodologies which are frequently used in science. They outline a neutral research approach that examines the facts or causes of any social phenomena in a systematic way. A positivistic approach identifies, measures, and evaluates any phenomena while supporting it with a reasonable explanation (Neville, 2007).
Using an abductive approach fits within the positivistic framework, and combines the deductive and inductive approach (Malhotra & Birks, 2007). With a deductive approach, the research shifts from general ideas and theories to specific and particular situations. Thus, the specific is figured out from the general. The opposite is true for the inductive approach, were the research proceeds from a particular situation to compose comprehensive general ideas and theories (Neville, 2007). This research shifts from the general idea of nostalgic advertising as an effective strategy to break through the advertising clutter, to the particular situation of positively influencing brand attitude and purchase intention among millennials, which directs the research in a deductive approach. Moreover, the extensive literature on nostalgic marketing and advertising suggest a solid foundation of the research. Hence, no literature known to the researchers, included millennials. Therefore, a deductive approach is not suitable. An inductive approach is also not sufficient due to the comprehensive literature available on nostalgia, brand attitude, and purchase intention. With this said, the abductive approach is considered the best fit for the research.
Strategies and choices
Saunders et al. (2009) state that it is important to choose an appropriate research strategy. The strategy should allow the achievement of the research and produce answers for the research questions. Since the realism is the chosen philosophy, strategies and choices are selected appropriately.
As the purpose of the study was both descriptive and explanatory, descripto-explanatory was selected as the nature of the research. According to Saunders et al. (2009), descriptive research is the precursor to explanatory research. Descriptive research identifies and classifies the elements or characteristics of the subject (Neville, 2007). It is also known as research aiming to generate a precise representation of persons, events, or situations (Saunders et al., 2009). Moreover, explanatory research is attempted when few or no former studies exist. The intention is to look for patterns, hypotheses, or ideas that can be tested and will form the basis for further research (Neville, 2007). Besides identifying managerial implications, the research sought to uncover the nostalgic advertising trend aimed at millennials. This study examined the relationship between nostalgia and millennials as well as the effectiveness of nostalgia in video advertisements on brand attitude and purchase intention among millennials. Hence, the literature required for this research was insufficient, meaning that there was no literature combining nostalgia with millennials. It can be concluded, that the nature of the research was descripto-explanatory.
A quantitative method, surveys, was used to collect, analyse, and summarise the data required to answer the research questions. Surveys structurally collect data from a relative large sample, with the data commonly presented as numbers. Moreover, a quantitative method is suitable to describe characteristics of a population or market (McGivern, 2013). This research analysed a trend in the market regarding a specific generation. Therefore, the use of this method was appropriate. Within surveys as research method, a questionnaire was used as research instrument to collect the quantitative data. More information about the survey can be found in §3.5.1 Survey design. With a single data collection technique, this research utilised a mono-method (Saunders et al., 2009).
Furthermore, the research includes primary and secondary data. The primary data was obtained from the questionnaire, whereas the secondary data was retrieved from reliable Internet sources, various academic articles, and books. This was acquired from the library’s databases of Jönköping University, Google Scholar, and other trustworthy search engines.
Time horizon
This study involves a homogenic group, specifically the employed millennials within the JIBS alumni network. The research examines if they have differences or similarities, namely if they are nostalgic at one particular time. The study is cross-sectional as it contains a close analysis of a situation at one particular point in time. Meaning that it is a ‘snap-shot’ result (Neville, 2007).
Techniques and procedures
In order to gain and process the data, techniques and procedures are undertaken. It includes a description of the survey design, target group and sampling process, and the data analysis design.
Survey design
This research utilised the quantitative data collection technique, surveys. Surveys include selecting a representative and unbiased sample of subject derived from the studied group (Neville, 2007). This study enclosed an analytical survey, because the relationship between different elements were analysed in the sample group (Neville, 2007).
As mentioned, an online questionnaire was used as research instrument. The questionnaire can be found in appendix 2. Qualtrics was used as program for the questionnaire design and data collection. The questionnaire was experimental, because the influence of nostalgic versus non-nostalgic advertisements was investigated. Respondents were randomly selected into either the experimental group or the control group. In the experimental group, the respondents are exposed to a nostalgic video advertisement whereas in the control group, the respondents are exposed to a non-nostalgic video advertisement. Both advertisements were from Microsoft which, as mentioned, has used nostalgic advertising to connect with their customers (Friedman, 2016). Also, the brand has existed for a long time, therefore, researchers assumed that consumers could be nostalgic towards that brand.
A pilot-test among five participants was held to test the questionnaire’s suitability and interpretation. After pilot-testing, some changes were made. Some items on the nostalgia measurement scale were perceived as difficult to interpret. Items were rewritten or replaced to avoid misleading interpretations. Also, to better test the opinion towards the advertisement, more items were added to the measurement scale. Regarding the question of respondent’s working situation, more options were added, such as ‘in between jobs’.
Target group and sampling process
The employed millennials within the JIBS alumni network were chosen as the target group, which was of various reasons. Firstly, it allowed for a homogenous sample, as the respondents had comparable educational backgrounds and shared a similar business perspective. Secondly, the likelihood of these millennials to be employed was high. Thirdly, the researched sought to analyse an ambitious dataset, which was achieved by targeting a large population.
The target group was reached through a combination of two non-probability sampling techniques: judgmental and convenience sampling. Non-probability sampling occurs when the researcher has little initial control over the selected respondents, or when it is not necessary to control respondents (Neville, 2007).
A judgemental sample allows using a judgement by the researchers to select respondents among criteria that are important for the research (Neuman, 2005). In this research, the criteria include millennials, consumers born between 1980 and 2000, who are part-time (non-student) or full-time employed. Additionally, the respondents were obtained based on convenience. The most convenient were sampled, also known as those immediately available (Neville, 2007).
The questionnaire was send out via e-mail to the target group by the coordinator of the JIBS alumni network. Therefore, the researchers had no control over the selected respondents within the target group. The e-mail (appendix 3) was sent out on March 22nd (2017) and was received by 3.437 e-mail addresses. A deadline for filling out the questionnaire was set on March 31st (2017) at midnight. Thus, the data was collected in ten days.
Data analysis design
The data retrieved from the online survey was quantitative data. Qualtrics processed the data from the questionnaire, after which the data was downloaded into the statistics software SPSS. The data was cleared by removing the uncompleted responses and not useful variables, such as IP address. Regarding the Likert-scale questions, reversed scaling was applied where it was required. To analyse the data and test the hypotheses, several statistical tests were executed with aid of SPSS. Besides, semantic differential graphs were created in the program Microsoft Excel to give better understanding on the analyses of hypotheses H3 (attractiveness), H4 (brand attitude), and H5 (purchase intention). The semantic differential graphs represent the weighted averages of each item of the measurement scales, which were calculated with the data from frequency distributions. The statistical tests are displayed in table 4.
Ethicality, reliability, and validity
To ensure that everyone involved in the research, including researchers, respondents, other users of the research, and the wider community, know what is and is not acceptable behaviour in the conduct of the research, the ethical standards are important (McGivern, 2013). Therefore, the MRS Code of Conduct was considered during this research. The Principles of the MRS Code of Conduct can be found in appendix 4.
Reliability
Reliability represents consistency, and therefore, it tests the strength of the questionnaire. It checks if coherent outcomes will be found with the same questionnaire at different times and under diverse circumstances, for instance dissimilar samples (Saunders et al., 2009). To ensure that the variables of nostalgia, attractiveness, brand attitude, and purchase intention were measured reliably, the researchers carefully selected suitable measurement scales for each variable. Moreover, the internal consistency of these measurement scales was statistically tested with the coefficient of reliability Cronbach’s alpha.
Validity
Internal validity refers to that researchers measure what they intend to measure. Thus, the findings of the online questionnaire needed to correlate with the reality (Saunders et al., 2009). Cooper and Schindler (2003) state, that to ensure a valid research, the following criteria should be fulfilled: content validity, criterion-related validity, and construct validity. Content validity is the validity of each measurement question or item and how well each measurement question covers the research question (Saunders et al., 2009). By conducting a thorough literature review, this type of validity is covered. All measurement scales used in the questionnaire were derived from solid literature, ensuring content validity. Moreover, criterion-related validity concerns the measurements’ questions predictive capabilities (Saunders et al., 2009). An overall correlation analysis tested this validity (appendix 5). Furthermore, construct validity covers to what extend the measurement questions measure the constructs which the researchers intended to measure (Saunders et al., 2009). This type of validity is also tested by use of the overall correlation analysis.
Summary
The empirical study of this research pursued a quantitative research approach. This research followed the realism philosophy, with a positivistic, abductive approach. Besides identifying managerial implications, the research sought to uncover the nostalgic advertising trend aimed at millennials. Therefore, the nature of the research is descripto-explanatory. By means of judgemental and convenient sampling procedures, the employed millennials within the JIBS alumni network were sampled for the online survey. The research provided a ‘snap-shot’ result, and was therefore a cross-sectional type of study. Primary as well as secondary data was obtained. Quality was ensured through the ethicality, reliability, and validity of this research.
Findings
The findings present the process of the analysis and the results obtained. Initially, the distribution of the dataset regarding experimental groups, gender, and age is outlined. Before testing the hypotheses, the outcomes of the reliability tests are explained. Finally, the five hypotheses are analysed, in which the results acquired describe whether the hypothesis is accepted or rejected. The statistical output of all tests can be retrieved from appendix 6.
Dataset
In total 381 responses were recorded, from which 115 uncompleted responses were deleted. From the remaining 266 responses, another 22 respondents, who were unemployed, were deleted as the research targeted employed millennials. The final dataset used for the analyses, consisted of 244 respondents.
Respondents were randomly assigned to either group A, who were exposed to the nostalgic video advertisement, or group B, who were exposed to the non-nostalgic video advertisement. The division between the two groups is equally distributed with 118 respondents being exposed to the nostalgic advertisement of Microsoft (Group A) and 126 respondents being exposed to the non-nostalgic advertisement of Microsoft (Group B). Table 5 displays the distribution of gender and age within each group as well as the total dataset.
Reliability
The Likert-scale questions (measurement scales) were tested on its internal consistency with coefficient of reliability Cronbach’s alpha. The measurement scales, including its items as well as the Cronbach’s alpha, are presented in table 6. Note that reversed scaling applied for items five, six, and seven of the nostalgia scale.
Referring to the nostalgia scale, the Cronbach’s alpha was originally .641, but as this is not highly favoured, one item was deleted to improve the Cronbach’s alpha to .661, and therefore obtain the best possible score to strengthen the internal consistency. This resulted in the deletion of the eighth item: ‘products from my childhood are pleasant reminders of my past’.
For evaluation of advertisement, brand attitude, and purchase intention, the Cronbach’s alpha value indicated a high level of internal consistency. Therefore, there was no need to delete items regarding these questions from the analyses of the dataset.
Hypotheses testing
To test the proposed hypotheses, various statistical tests were conducted. The mean (average) scores for the nostalgia scale, evaluation of advertisement, brand attitude, and purchase intention, were measured to conduct the appropriate analyses.
Millennials and nostalgia
This section outlines if H1: millennials are nostalgic towards their own personal past, can be accepted. To display if the respondents are nostalgic, a frequency distribution was carried out. It showed that most respondents sometimes ‘think about their childhood’. Only 35 respondents stated that they never or rarely ‘think about their childhood’. With 244 respondents in total, it can be concluded most respondents ‘think about their childhood’. Furthermore, almost all respondents state that they can recall pleasant memories from their childhood. Moreover, several items were tested to investigate if the respondents are nostalgic. The initial outcome of the frequency distribution suggest that the respondents are nostalgic towards their personal past.
Additionally, the strength and direction of the linear relationship between frequency of thinking about childhood and nostalgia was tested with a correlation analysis. Table 7 presents the Pearson Correlation and significance value on both variables.
After the correlation analysis was carried out, the number of cases had to be checked on correctness. This ensured that there was no data (respondents) missing in the analysis. The number of cases had to be equal to the amount of responses in the dataset. In this research, the number of cases were correct. Thereafter, the relationship between the variables had to be considered. This means noticing if there is a negative sign in front of the r value, because if there is, there is a negative correlation between the two variables. Since the outcomes of the correlation analysis showed no negative sign before the r value, the variables were positive. This indicates a positive correlation between frequency of thinking about childhood and nostalgia. Thus, the more the respondent thinks about their childhood, the more nostalgic they are. Moreover, the size of the value of Pearson correlation had to be above zero, to assure a relationship between the variables. The Person correlation value also indicated the strength of the relationship between the variables. As the value of the correlation was above zero, namely .322, it can be concluded that there is an intermediate relationship between the variables frequency of thinking about childhood and nostalgia. The significance level of the correlation analysis is .000<.05., this means that H1: millennials are nostalgic towards their own personal past is accepted. Thus, millennials are nostalgic towards their own personal past.
The correlation was also tested among the same variables with various selected numbers of cases in the dataset. By doing so, the stability of the results was investigated. If the results remained the same among the different selected groups from the dataset, the outcomes are generalizable. Meaning that, when the respondents of the dataset change, for example increase with 50 respondents, the outcomes will remain the same. The different numbers of cases in the dataset selected included: only females, only males, and a random selection of 50 per cent of the respondents. Among all these cases, the test outcomes showed a significance level smaller than .05 as well as a positive correlation. Thus, the results of the correlation analysis were stable. With different respondents, there will not be a difference in the outcome. This makes the acceptation of H1 stronger.
Age of millennials and nostalgia proneness
Two crosstabulation including Chi-Square tests were conducted to test H2: the age of a millennial influences the nostalgia proneness. The crosstabulation displayed the observed and expected count, which showed similar outcomes. Thus, there is no different experience between the observed and expected count. The first Chi-Square test determined if year of birth and nostalgic feelings were related. The second Chi-Square test determined if year of birth and thinking about childhood were related. It is assumed by the researchers of this study, that if the respondents think frequently about their childhood, they are nostalgic towards their own personal past. Both outcomes of the Chi-Square tests are summarised in table 8.
Referring to year of birth and nostalgic feelings, the corrected value is 334.328, with respectively p(.948)>.05, meaning that there is not a statistically significance. This implies that the difference between younger and older millennials regarding the nostalgic feelings is not significant.
Concerning year of birth and thinking about childhood, the corrected value is 53.641, with respectively p(.565)>.05, meaning that there is not a statistically significance. This implies that there is no significant difference between younger and older millennials regarding the frequency of thinking about their childhood.
Both Chi-Square tests do not show a statistically significance. Therefore, H2: the age of a millennial influences the nostalgia proneness is rejected. In the millennials generation, there are no differences between age and the intensity of nostalgic feelings.
Attractiveness of nostalgic and non-nostalgic advertisement
The respondents were asked to evaluate the advertisment by comparing both types of video advertisments on five items including unappealing-appealing, bad-good, unpleasant-pleasant, boring-interesting, and unlikable-likeable. The attractiveness was analysed by means of a semantic differentials graph. It signified the attitude respondents had towards the video advertisement. Looking at the semantic differential graph (graph 1), the nostalgic video advertisement of Microsoft was perceived as highly attractive. The majority of the respondents in group A agreed, to a great extent, that the advertisement was appealing, good, pleasant, interesting, and likeable. The non-nostalgic advertisement was also perceived as attractive, but when comparing both type of advertisement, the nostalgic advertisement outperformed the non-nostalgic advertisement.
To test the effect of nostalgic cues in advertisements on the attitude towards the video advertisement among millennials, an independent-samples t-test was conducted. It indicated whether there was a significant difference between the two types of video advertisements and their attractiveness, by comparing the mean scores of the nostalgic and non-nostalgic video advertisement. It also enabled the researchers to test whether H3: the nostalgic video advertisement is more attractive compared to the non-nostalgic video advertisement should be accepted or rejected. The output is shown in table 9.
With Sig.(.137)>.05 in Levene’s Test for Equality of variances, equal variances were assumed. With respectively, p(.000)<.05, it is implied that there is a significant difference in the attractivess between the nostalgic and non-nostalgic video advertisements. The respondents showed distinctive attitudes towards the two different types of video advertisements. Which advertisement is deemed more attractive as well as having a more favourable attitude towards it, is decided by comparing their mean scores. With the nostalgic video’s mean score of 5.59>4.88, compared to the mean score of the non-nostalgic video, the nostalgic video is evaluated as more attractive. This results in accepting H3: the nostalgic video advertisement is more attractive compared to the non-nostalgic video advertisement. The type of video advertisement has a positive effect on the attractiveness, meaning that the nostalgic video advertising is more attractive compared to non-nostalgic video advertising. Millennials have a more favourable attitude towards the nostalgic video advertisement compared to the non-nostalgic video advertisement.
Table of Contents
1 Introduction
1.1 Background
1.2 Problem definition
1.3 Research questions
1.4 Delimitation
1.5 Contribution
1.6 Limitations
1.7 Definition of key terms
2 Literature review
2.1 Nostalgia
2.2 Nostalgia marketing
2.3 Nostalgic advertising
2.4 The effects of nostalgic advertising on brand attitude and purchase intention
2.5 Characteristics and behavioural patterns of millennials
2.6 Hypotheses and conceptual model
2.7 Summary
3 Methodology
3.1 Research philosophy
3.2 Research approach
3.3 Strategies and choices
3.4 Time horizon
3.5 Techniques and procedures
3.6 Ethicality, reliability, and validity
3.7 Summary
4 Findings
4.1 Dataset
4.2 Reliability
4.3 Hypotheses testing
5 Analysis and Discussion
5.1 Relation nostalgia and millennials
5.2 Effect of nostalgic advertising on brand attitude and purchase intention
5.3 Summary
6 Conclusion
6.1 Research conclusions
6.2 Managerial implications
6.3 Ethical and social impact of findings
6.4 Future research
7 References
GET THE COMPLETE PROJECT
