Inductive and deductive approach
The most traditional approach when conducting a research is a deductive approach; where the research develops from being general to specific (Hornig Priest, 1996). The research begins with theory as the general aspect and connects this to a specific case where hypothe-sis will be tested. Inductive approach, on the other hand, moves from a specific case with the purpose of creating a common theoretical framework (Hornig Priest, 1996).
When performing an ethnographic study, new data is gathered from research and conclu-sions are made from the findings in order to create new theory, or to identify relationships between unforeseen factors (Saunders, Lewis & Thornhill, 2007). This approach is induc-tive, i.e. where new theory is built. According to Machin (2002), this method is mostly used for ethnographic studies. Despite that inductive studies do not require knowledge of theory prior to the study; the authors have decided to include a chapter presenting previous re-search. This chapter is not seen as theory, but more as a clarification of the existing con-cepts, such as multitasking.
Qualitative and quantitative data
Hornig Priest (1996) defines qualitative research as in-depth understanding of human be-havior. The researchers aim at understanding, exploring, and identifying underlying factors of behavior and answering the question why some behavior occur and create a framework. Quantitative studies, on the other hand, focus on making generalizations and a representa-tive result for the designed populations (Marshall, 1996).
Since the purpose of this thesis is to identify behavior, qualitative research is an appropriate tool to gather the primary data. Moreover, previous research has mostly been quantitative; in order to contribute, new research using a qualitative methodology where behavior is identified could be beneficial. In order to obtain background to the problem, the collection of secondary data will be research concerning people‘s behavior when watching TV, e.g. TV viewing statistics. This will be useful to introduce the field of research in order to set the actual time for the observation, while behavioral research helps the observation and helps the process of identifying important data. It is also useful to test certain theories in the field and see if they are accurate or not.
Primary and secondary data
Data collection is divided into primary and secondary information. Primary data is informa-tion collected by the researcher from firsthand experience such as interviews, observations and surveys. The primary data method is aiming at answering a purpose specific for the study; it is therefore highly reliable due to its distinctiveness (Collis & Hussey, 2003). On the contrary, secondary data is data that was gathered for previous purpose by someone other than the researcher and has already been published. It has a wider variety of sources than primary data, e.g. research papers, articles and textbooks (Collis & Hussey, 2003).
In order to answer the purpose of this report, both primary and secondary data will be used. Secondary data will be based on previous research, and used to provide background information to the problem. Primary data will be of main focus in this report and will con-sist of an ethnographic case study where four independent participants in the ages of 20-25 will be observed in their home environment when watching TV.
Sampling and selection
In this ethnographic study, the researchers have decided to, through participant observa-tions, observe two men and two women within the age of 20-25. All participants are stu-dents living in Sweden. The reason for the selection is practical reasons such as economic, time-wise and possible access to people‘s living rooms. The participants were selected based on their type of lifestyle, i.e. having different level of activities in their everyday life, interest and preferences. As well, the researchers wanted two men and two women in order to provide information of how each gender behaves. By choosing participants amongst people known to the researchers, the selections were based on who would help to give an as accurate picture of the target group as possible. The observations will take place in the informants‘ own apartments during 10 sessions per person.
The selected sample for this case study is non-random since it is chosen by the researchers. It is argued that random sampling probably will result in a representative sample if the cha-racteristics of the research are normally distributed for the population (Marshall, 1996). However, the aim is to understand attitudes and beliefs through a qualitative investigation and there is no proof that they are normally distributed, which makes the probability ap-proach inappropriate for this study (Marshall, 1996). Small samples can provide just as in-formative data and it is important to know that generalizability is not the aim of this research (Marshall, 1996). The appropriate sample size for this study is the one that will answer the questions asked. Therefore, the authors believe that a sample of four persons is large enough for this case study. Since the persons are selected in order to make an as productive sample as possible it is called a judgement sample (Marshall, 1996). Selection is based on sim-ple demographics such as age, gender and occupation but also whether they will be able to provide the desired information. For instance, if a person does not own a computer or TV, there would be no point of including them. The judgement sample for these four cases was based on certain requirements that would make the four cases representative.
Further, the researchers argue that there are three advantages with the participants being known to the observers. First of all, it ensures that they actually own a TV and fulfill other requirements for taking part in the study. Secondly, since the observation takes place in their home environment and they know the observer, it allows them to be more comforta-ble than having an unknown person observing them. There is no awkward situation of a stranger in the room and they can behave as natural as possible.
The researchers also argue that knowing the participant also benefits the analysis. It is easi-er to understand why a person reacts in a specific way or notice different things if their pre-ferences are known to the observer. By this, one can provide information about characte-ristics and preferences, such as ―I enjoy cooking and traveling‖. However, there is a possi-bility that important aspects are left out unconsciously. Then, if the observer knows that this person also is passionate about animals, even if it is not mentioned, more information can be analyzed and understood better. However, knowing the participants have its down-sides as well, for instance that certain patterns might be overlooked. If the observer is fa-miliar to the participant acting in a specific way it could be harder to notice as a specific behavior, which an unfamiliar observer might have noticed. Also, the researchers argue that the participants feel more comfortable with a familiar person observing. Yet they might still act differently, being concerned that the observer could judge them based on the behavior and still act differently than normal. It could be difficult to maintain the somewhat formal atmosphere, having to act differently than usual, e.g. not being allowed to talk to each oth-er. Keeping this in mind, the researchers argue that the positive aspects outweigh the draw-backs.
1.2 Problem area
1.4 Research questions
1.7 Structure of thesis
2 Previous research
2.1 Advertisement avoidance behavior
3.1 Primary and secondary data
3.2 Qualitative and quantitative data
3.3 Inductive and deductive approach
3.4 Sampling and selection
3.5 Research strategy: Ethnography
3.6 Data collection
3.7 Grounded theory
3.8 Practice theory
4 Empirical findings
4.1 Case 1 – Emma
4.2 Case 2 – Johan
4.3 Case 3 – Anna
4.4 Case 4 – Andrea
4.5 Summary of empirical data
5 Frame of reference
5.1 Attention gainers
5.2 Consumer media consumption model
GET THE COMPLETE PROJECT
Stay tuned! TV-commercial avoidance in a multitasking environment