Assessing Research Philosophy and Approach principles

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

The research philosophy which I will undertake for my research study will be that of a positivist approach. The reason for this chosen approach is that there is already previous theoretical framework in to the field that I am looking in to from which I can basis my research direction. Remenyi et al, (1998 in Saunders et al, 2003:83) states that the positivist approach is based on "working with an observable social reality and that the end product of such research can be law-like generalisations". Although the end product of the research is unlikely to reach the magnitude of "law-like generalisations" due to the size and scale of my research, however the study will be aimed at revealing new insights into the airline industry taking into account the economic downturn which has not been previously looked at extensively due to its recently occurring . The philosophical ideals of positivist research as shown by Remenyi help justify the suitability of the positivist approach for my research.

The research approach I am going to use is that of a deduction, Malhotra and Birks (2003: 141) stated that within a deductive approach "the issues to focus an enquiry upon emerge from the established theoretical framework".

Due to the relatively new elements that must be taken into account in undertaking my research into the airline industry such as the currently volatile and changing global economy which is currently recovering from a recession. The focus on "testing from theory in a new context" (Malhotra et al 2003: 141) will allow more focus to further previous research and to take into account new elements and variables such as the recession and new regulations.

The deduction process works on list of "five sequential stages" as mentioned by Robson (2002 in Saunders et al, 2003:117) which mark the way in which the research will progress:

Deducting a hypothesis

Expressing the hypothesis in operational terms - this is an indication of exactly how concepts or variables that are posed will be measured.

Testing the hypothesis

Examining the outcome of the research - this will either confirm the theory or show other results which may indicate a need for further research and modification of the research in future testing.

If needed modify the theory in light of findings.

The five sequential stages of deductive research (Robson 2002 in Saunders et al, 2003:117)

As shown in the stages above, the deductive research approach also leaves scope for future exploration and research into the area. This is key in the area of the airline industry, due to the changes and structural rearrangements brought about through deregulation ,and the recent economic . Further research may be need in future years to measure any further changes or structural shifts within the market.

3) Time Frame

Time-constraints are one of the major issues in carrying out this research. Due to the limited time I have to complete my research study I will be adopting a cross-sectional study. A cross sectional study as defined in Proctor (2005: 554) is one which uses "a number of different respondents at single point of time". This allows me to gain a snapshot of the market as it currently stands which is ideal for the time restraints I am currently under.

Another factor which has lead me to choose a cross-sectional approach is as Saunders (2003) mentions these types of surveys can be focused on taking into account a particular phenomenon. This fits in well with the economic recession (something which has not been experienced since 1992) and how the airline industry is recovering from it.

4) Quantitative Approach

The use of quantitative data is now a featured norm in viewing a larger audience and generating a strong factual based argument to be generated through statistical analysis (Brassington, 2006). Wright et al (2000: 19) points out how the strength of quantitative research lies in the way it uses "mathematical analysis and modelling to explain marketing phenomena". If this is combined with a large enough representative sample then it is possible for the researcher to apply the results to the wider population (Brassington, 2006)

However it is mentioned that quantitative data research "scrapes the surfaces of people's attitudes and feelings" (Wright et al 2000: 19). This points to the idea that quantitative data lacks the deep in depth interpretation of interviewee based qualitative data. However, due to the costs and time restraints surrounding the use of qualitative research techniques it is not always viable for research projects.

The reasoning for a quantitative approach to this research project is as follows:

Accessibility to target sample: The sample base for the project is based around those working in the business sector that fly for their business. This means that it is hard to get enough time with them for a meaningful in-depth interview, therefore the use of email questionnaires based around a quantitative question base is better suited for reaching the target sample.

Sample size needed: Although the target sample allows for a smaller target sample than other studies, a suitably representative sample size will be required, which lends itself more to a quantitative approach as it will require the researcher to distribute the data collection tool to a suitably wide sample base.

Evolution of distribution technologies: the revolution of the internet based questionnaires, allows much wider distribution ,more efficiently collection, and more integration of all data sources, for both online and offline (Brassington, 2006)

5) Reliability and Validity of Research


As defined by Bryman et al (2003: 33) "reliability is concerned with the question of whether the results of a study are repeatable". Saunders (2007) points out that questionnaire reliability is something that must be focused upon whether a questionnaire can, under different conditions at differing points of time, produce consistent results. Mitchell (1996 found in Saunders 2007: 367) identifies three approaches that can be used to assess a questionnaires reliability:

Test re-test - Repeating the same questionnaire with the same respondents at differing times.

Internal consistency - Involves comparing responses to each question in the questionnaire with those to other questions in the questionnaire (Saunders et al, 2007: 367)

Alternative form - This involves comparing responses to alternative forms of the same question or groups of questions. (Saunders et al, 2007: 367)

However, ensuring the questionnaire is completely valid is extremely difficult as the ability to either replicate the same response conditions for the respondent, or following the path of response comparison to a high enough level to ensure reliability ,involves a high level of skills and difficulty for the researcher.


As defined by Saunders (2007: 366) "validity is the ability of your questionnaire to measure what you intended it to measure". Validity is distinguished into four main types:

Measurement validity

Internal validity

External validity

Ecological validity

(Bryman et al, 2003)

This is an important feature of a quantitative questionnaire as it can greatly affect the usefulness of the results once the data has been collected. The questions asked aren't focused correctly and do not measure what they are intended to measure whether that be through incorrect wording for example, then the data may lack any real usefulness to the researcher as it will not meet the original needs of the research.

6) Research Methods Utilised

Internet based questionnaires (embedded in an email)

The main benefit of an interview based questionnaire to this research is the ability to access the target sample through a highly used media of the target audience, in this case email. Although response rates for email based questionnaires as mentioned by Bradley (2007) were around 10% in 2003, due to the spam element of email marketing and research which has developed in parallel with the internet explosion, in May 2009 95% of all global email messages were spam (, 2010). However, a higher rate of response is expected due to senior contacts within the business being used as the target sample.

The global interconnectivity of the internet reduces the time constraints forced upon the research process, as email based questionnaires can reach many global respondents instantly and responses can be returned at the same speed to a singular collection point, it is ideally suited for focusing on a sample which is fast moving and may be located over a globally geographic area. This ease of distribution and collection methods also falls into line with the relatively non-intrusive nature of internet interviewing, in that internet questionnaires are "quick and easy to prepare, administer and analyse , as well as being relatively low-cost" (Bradley, 2007: 135). These factors make this questionnaire type perfect for a cross-sectional study which focuses on fast-paced, evolving market, in a global ly distributed and fast-moving sample.

The table below from Saunders et al (2007: 358) shows the attributes of internet questionnaires:


Internet Questionnaire

Population's characteristics for which suitable

Computer-literate individuals who can be contacted by email, internet or intranet

Confidence that right person has responded

High if using email

Likelihood of contamination or distortion of respondent's answer


Size of sample

Large, can be geographically dispersed

Likely response rate

Variable, 30% reasonable with organisations/via intranet, 11% using internet

Feasible length of questionnaire

Conflicting advice; however, fewer 'screens' probably better

Suitable types of question

Closed questions but not too complex, complicated sequencing is fine due to its use of IT, must be of interest to respondent

Time taken to complete collection

2-6 weeks from distribution (dependent on number of follow-ups)

Main financial resource implications

Web page design, although automated expert systems offered online and by software providers are reducing this dramatically

Role of the interviewer / field worker


Data input

Usually automated

(Adapted from Saunders et al 2007: 358)

6.1) Strengths and weaknesses of internet questionnaires outlined


Wider question formatting options, e.g. pull-down menus

Can decide on question view seen by respondent, i.e. the number of questions viewed at one time on the screen (Bryman et al, 2003)

The ability to filter questions, whether it be to filter respondents out of the questionnaire who don't fit the sample criteria or only allowing certain sample respondents to answer a certain section/s of the questionnaire.


In some cases unintentional respondents may answer the questionnaire.

Apart from sending follow up emails there is no way to ensure questionnaire is carried out.

If questionnaire is not constructed to a high enough / suitable level there may be question interpretation problems, leading to incorrect results being received.

7) Questionnaire Design and Content

In order to develop a successful and comprehensive questionnaire Aaker et al (2001) suggests that using the following five point plan will help the completion of an effective questionnaire. The questionnaire design and content process is essential, as a poor questionnaire design that lacks focus or judgement may diminish the value of the survey. This is the five point plan laid out by Aaker et al (2001):

Planning what to measure - Deciding what is to be asked under the research issue

Formatting the questionnaire - Decide on the format of each question

Question wording - Determine how the question/s will be worded

Sequencing and layout decisions - Layout of questionnaires and grouping questions

Pretesting and correcting problems - Check questionnaire and make any changes

Adapted from Adapted from Aaker et al (2001: 304)

The structure and sections chosen for the questionnaire used in this study can be seen below:

Questionnaire Sections

Description of Section

Section 1: Air Travel

This section is the initial section of questions relating to peoples air travel information. The section is aimed at filtering out respondents and also gaining responses to be used across tabulator factors when it comes to the analysis of the data.

Section 2: Flying for Business

This section is focused on people who fly or have flown for their business. It is aimed at seeing the factors involved when flying for business purposes. This can draw a comparison to see whether the economy may have affected a business's flying policy.

Section 3: Personal Flying

This middle section is based on personal flying preferences and the respondents' feelings towards different areas of the airline industry, i.e. short-haul flying, long-haul flying, legacy airlines, etc. In this section the majority of the questions use Likert scale, these scales are used to reduced bias and get a fairer rating of the elements looked at in each question.

Section 4: The Economy and Flying

This area of the questionnaire is based on a number of statements about flying and the economy, by using a Likert scale judging how much the respondent agrees or disagrees with the statement. The aim of this section is to see how much the respondent consciously views how the recent economic crisis has affected theirs and other decision to fly.

Section 5: About You

This section is focused on demographic questions which can be used against previous question responses to identify any trends which may occur.

8) Sampling

The target population for my survey is that of people in the business sector who preferably fly for their company. Due to the time constraints and the large population of people who fly, it was important to focus the sample group. In order to reduce the potential size of the sample and to gain a more focused argument the research will focus on a particular group, in this case people that use the airline industry and who operate within the business sector.

To gain a fuller picture of this target population the researcher will utilise contacts within a business in this case Aker Solutions to distribute the questionnaire. The questionnaire will be distributed throughout the business with the aim to get a representative sample of the overall business community.

This utilises the cluster sampling theory, this type of sampling "can be based on any naturally occurring grouping" (Saunders et al, 2007:223), in the case of the research the cluster sample is those of people who operate within the business sector , utilise air travel and work for the chosen respondent company. Although the use of cluster samples has its restrictions with Bradley (2007: 175) mentioning that it may not give a true picture of the population. However due to the availability of the particular sample, the opportunity to gain a sample base in the form of a population cluster such as the one used in the research, has allowed the research to be carried out successfully in a short period of time.

As mentioned in Saunders (2007) there are a number of factors that must be considered when deciding on a suitable sample size:

The confidence you need have in your data - that is, the level of certainty that the characteristics of the data collected will represent the characteristics of the total population.

The margin of error that you can tolerate - that is, the accuracy you require for any estimates made from your sample.

The types of analyses you are going to undertake - in particular the number of categories into which you wish to subdivide your data, as many statistical techniques have a minimum threshold of data cases for each cell.

A sample size of 30 to 35 responses was deemed an acceptable number in order to analyse the data to an extent that meaningful conclusions could be drawn from them. 41 results were completed from emails sent out which is more than the initial response rate set out.

9) Data Analysis

In order to analyse the data gathered from the questionnaires, it will be entered into SPSS software for analysis. The decision to use the SPSS software is due to its ability to comparing and analysis large amounts of data. With the ability to analyse two variables against each other at once there is the possibility for much more in-depth analytical outcomes, which will help prove any hypothesis.

There is also the element that SPSS can produce transferable data tables (see appendices) which can be shown as tables or transferred into other products to create tables and graphs which can clearly demonstrate any trends or anomalies which may occur within the data collected.