Principles of Business Data Analysis 2024/2025 October 6 University PDF

October 6 University Faculty of Tourism & Hotels Bachelor Degree Programme Tourism and Leisure Management (October 6 University (Egypt) & IMC Krems (Austria) Principles of Business Data Analysis...

October 6 University Faculty of Tourism & Hotels Bachelor Degree Programme Tourism and Leisure Management (October 6 University (Egypt) & IMC Krems (Austria) Principles of Business Data Analysis Prof. Dr. Hanaa Fayed 2nd Year - 2024/2025  Secondary Data Analysis in Lecture 1 Leisure and Tourism Research  At the end of this lecture you will be able to:  appreciate the differences between primary and secondary data Learnin  understand the difference between secondary data collection and secondary data analysis g  identify some of the major sources of leisure and tourism data Outcom  acknowledge the potential of the Internet and AI as a data source es  assess the advantages and limitations of conducting secondary data analysis  provide an evaluation of secondary data analysis What is secondary data analysis? 1. The phrase refers to the analysis of information collected for a purpose other than that of the researcher - in this sense the researcher becomes the secondary user of the data. 2. We can contrast this with primary data, which is original data generated by-new-research-using techniques such as surveys, interviews or observations. 3. Most research will, of course, include an element of secondary data collection to discover what work has already been carried out on a particular subject. 4. This is a necessary first step in any research design, and forms an important part of the literature review stage.  What is secondary data analysis? 5. An important point - you should always consult- secondary sources of information-before-you-start collecting-primary data as you may well find that the information you need is already there. 6. A great deal of time, effort and money can therefore be saved if you are aware of available data, and where to look for it. 7. So, secondary data-collection should always come before primary data collection!  Understand the difference Lecture 2 between Data collection and Data analysis. 1. The word 'analysis' in the title is the key one, When carrying out this type of research, you are not just collecting information and reproducing it in similar or identical form; you are, instead, re-working the data to address your own research objectives. 2. Hakim (1982: 1) defines secondary data analysis as 'any further analysis of an existing dataset which presents interpretations, conclusions, or knowledge additional to, or different from, those presented in the first report on the inquiry as a whole and its main results. 3. This method usually refers to the analysis of quantitative data generated usually by surveys. 4. The term secondary analysis can, however, also be applied to the analysis of diaries, letters, reports, television and radio broadcasts etc, in this case it refers to qualitative data analysis. The research associated with these sources of qualitative information is more usually referred to as content analysis. it is important to appreciate that secondary data analysis can be carried out on both quantitative and qualitative data.  The potential of secondary data analysis in leisure and tourism research   Impacts Relationships marketing strategy tourism development comparison track changes  We can use secondary data analysis to looking at the impacts of a particular tourism legislation, or the impact of a new policy for tourism, or the impact of a specific event. For example, what has been the impact of the departure tax on flights from Egypt? What impact did a particular terrorist attack have on tourism in that area?  We can explore relationships through the use of secondary data analysis. For example, is there a relationship between social class and patterns of tourist activity? Is there any relationships between gender, age and socio-economic grouping and types of tourism activities  We can determine the activity and tourism types for every distention using secondary data. For example, you could look at areas of heritage tourism activity, diving and desert tourism to plan for a marketing campaign for these areas  You could use secondary data for area study, and also investigate penetration rates to see what proportion of target market segments is being attracted to the facility.  A feasibility study for a new leisure facility could be undertaken by combining secodary data with participation rate data from a national survey.  We can make international comparisons through secondary data analysis. For example, you could compare levels and trends in second-break tourism in England with those in Germany.  We can look at the use of time-budget research to explore how people classify and use their leisure time, and therefore track changes in the use of time over a certain period.  Secondary data Lecture 3 for leisure and tourism   We live in the information age. Information never goes away, it just gets added to, and at an exponential rate. The amount of information out Secondary data there, if only we knew where it all was and had the time and money to access it all, is frighteningly for leisure and large. tourism  You cannot possibly become familiar with all available sources of information. Your task as a leisure and tourism researcher is to familiarize yourself with the most important and accessible sources of data.  Most of the available data is quantitative in nature i.e. details of tourist numbers, tourist nights, tourist expenditure and leisure participation rates.   Sources of such data include international bodies, central and local government, non-governmental organizations and commercial organizations. Also books, articles and reports from scientific journal and Sources date base engine. Secondary data for leisure and tourism  Using the Internet as a source of secondary data The Internet is a network of computer networks linking millions of computers around the world. The World Wide Web (WWW) is part of the Internet network with web pages organized into websites. The address of a website is known as a Universal Resource Locator (URL).  The advantages of secondary data analysis 1) forces the researcher to think more closely about the theoretical aims and substantive issues of the study. 2) Shifts the focus from individual data subjects to a broader analysis of social conditions and change. 3) Allows you to merge data from various sources in order to provide larger and more useable data sets. 4) Different individual data sets can also be used to test the same hypothesis thus establishing reliability if the same results are achieved.  The advantages of secondary data analysis 5) Requires less time and effort to collect the data. 6) The secondary data analyst can devote more of both to analysis and interpretation. 7) Offers a more flexible approach in that, subject to any deadlines, it can usually be carried out as and when it suits the researcher, and over long time periods. 8) The data covers long time periods. Lecture 4 Collection methods  Collection methods Tourism statistics are thus collected in a variety of ways around the globe. In some countries it is a legal requirement for accommodation businesses to collect data, in other countries statistics are generated by border crossings either through a process of registration or by a sample survey. As Latham (1998: 46) reminds us, 'Methodologies range from the highly scientific and sophisticated to the clumsy and naïve'. Given these problems of definition and differences in data collection methods,  Collection methods The data collection methods may then change over time. Let us take a leisure example from Sharm Elshikh, and compare participation in leisure activities between 2015 and 2020: In 2015, 38 per cent of respondents listed diving as one of their leisure activities. In 2020, this figure had risen to 57 per cent. This remarkable increase could be explained by aggressive marketing or by some other phenomenon. A more likely explanation is that in 2020 the interviewers prompted the respondents by mentioning certain leisure activities, one of which was diving. A Similar pattern was in fact found for the other 'prompted' activities. In other words, the method of data collection  1. Test the notion that participation in leisure activities increased dramatically in Egypt over the period 1980-96. Activity 1 2. Investigate the hypothesis that the type and intensity of leisure activity varies according to gender, age and socio- economic grouping.  1. Test the hypothesis that between 1980 and 1995 visitors to Egypt 'heritage' attractions increased at the expense of seaside areas. 2. Investigate the hypothesis that Activity 2 between 2020 and 2022 there was a growing East/South divide in the Egypt in terms of tourism numbers. Lecture 5 Qualitative Methods  Learning Outcomes At the end of this chapter you will be able to:  make an informed choice on how to collect qualitative data.  assess the strengths and weaknesses of research using qualitative methods.  distinguish between different types of interviewing, including focus groups.  assess the value of methods of data collection.  use a case study strategy to research. The Case for qualitative methods: Scott and Godbey's paper (1990: 194-9) discusses how qualitative research can be used to study leisure behaviour: 1. Leisure as experience 2. Leisure behaviour as a formative process - the form and meaning of leisure involvement changes over time as new situations and circumstances are confronted. 3. Leisure behaviour as a group phenomenon - the individual can be analysed in the context of a group as leisure involvement is embedded in social interaction. Methods of data collection: Methods of data collection: Lecture 6 Interviewing  Interviewing Choosing to collect data using an interview needs to be thought out carefully as there are different types of interviews depending on the aims of the research. We can identify three main types of interviews defined by their degree of structure: 1. Structured interview: This type of interview is associated with the survey style of research where a standard interview schedule is designed to answer a series of specific questions on a face-to-face basis. This interview structure will produce quantitative data for analysis.  Interviewing 2. Semi-structured interview: Interviews of this nature will have specified questions but will allow more probing to seek clarification and elab- oration. They would have more latitude than the structured interview. 3. Unstructured interview: The name 'unstructured' is a misnomer as no interview can have a total lack of structure. This type of interview has also been called a 'semi-structured' interview (using the same name as the previous type), a 'depth interview, a 'qualitative' interview or an 'exploratory' interview. This type of interview is the one associated with ethno- graphic research where the aim is to understand the perspective of the interviewee and the meanings that the interviewee attaches to situations and contexts important to him or her. Lecture 7 Social surveys Learning Outcomes  At the end of this lecture you will be able to:  determine the appropriateness of the survey as a research design  demonstrate the importance of questionnaire layout, structure and presentation  design reliable and valid questions  utilize a variety of question types and attitudinal scales  draft a short questionnaire  assess the strengths and weaknesses of the different approaches to survey research Introduction  In our experience the survey method is one of the most frequently utilized designs in under- graduate dissertations in leisure and tourism.  Survey research involves asking participants direct questions through the following: face-to- face interview, telephone interview or post (self-administration). Gogol form.  The questionnaire is the normal survey tool, which is a series of printed questions schedule of some sort.  The purpose of the questionnaire is to obtain reliable and valid data on the subject being researched. Questionnaires and interviews are the prevalent methods of data collection in this research design but they are also used within experimental and case study designs.  Population: A key objective of survey research is to obtain data which is presentative of the population. In other words, research based on surveys is usually used to generalize from the sample to a larger population  consequently the issue of sampling is an important one.  A survey involves 1. collection of data (invariably, but not exclusively, by self-administered questionnaire, structured or semi- structured interview) 2. a given set of units (people or organizations) 3. a snapshot at a single moment of time (may be repeated over a given time interval) 4. Systematically obtaining quantifiable data on pre- determined variables, which are then analyzed. Stages in survey research 1. Appropriate conceptualization and structuring of the research problem 2. Derivation of appropriate measures of the key concepts 3. Determination of the sampling strategy 4. Construction of the questionnaire/interview schedule 5. Pre-testing the survey instrument 6. Piloting the survey 7. Refining and modifying the instrument and implementation process 8. Administration of the questionnaire 9. Data coding and processing 10. Data analysis 11. Report writing.  Techniques Lecture 8 of data collection  Face-to-face interviews face-to-face interviews are essentially structured conversations, or question and answer sessions. The conversation is structured by a schedule of questions which is administered by an interviewer to every respondent in the same way. The face-to-face interview, in its structured form, has one main advantage which is high response rates. The interview can introduce many sources of bias into the research and thereby undermine the reliability of the research instrument. The respondent will make available only that information which they think will be of interest.  Interviewing tips  Practice the interview beforehand.  Prepare a standard introduction explaining the purpose of the questionnaire and ensuring confidentiality. Make sure you have an identity card and a covering letter if appropriate.  Make sure you have sufficient questionnaires, pens, etc.  Dress appropriately  Inform the participant how long it will take be pessimistic.  Treat respondents with care.  Do Not appear judgmental about the responses-be careful about your body language too.  Follow the schedule and instructions systematically and closely  You should not interview children without the permission of their parents.  Always thank the respondents for participating   Postal and  Distributing googol form questionnaires bysurveys post or googol form is a kind of self administered survey, since respondents complete the questionnaire on their own  The postal and googol form survey are relatively inexpensive, not very time- consuming and often a highly effective and quick means of reaching a specific sample.  The self-completion questionnaire has the advantage that it may be completed In privacy, but the likelihood of misunderstandings and incomplete  Tips for postal and googel questionnaires  Make sure that the questionnaire has been pre-tested and piloted  Make sure the questionnaire is 'professionally' presented.  Consider pre-notifying the participants.  Construct an informative but brief covering letter  Prepare and implement a reminder as responses begin to wane.  Consider sending a second reminder and questionnaire to boost the response rate Lecture 9 Telephone surveys Telephone surveys The telephone survey is an alternative survey method, generally they have the advantages of being cheap, the physical appearance of the inter viewer does not matter and the potential sample size is. However, the response rate can be low and it is only suited to a small number of questions; certainly the interview should last no longer than 15 minutes! The design of the questions involves a staged process: 1. define the concept 2. break this concept down into a number of dimensions 3. develop indicators for each dimension 4. select one (or more) indicators for each dimension 5. design questions to collect information for each indicator 6. pre-test the questions in order to ensure that they are valid and reliable 7. the resultant questions form the variables in your data analysis Question types When constructing a questionnaire the flow and format of the questions is important. There are basically two types of question in questionnaires -open and closed questions. Closed questions have pre-coded answers, whereas in open questions respondents are encouraged to express themselves more freely. Question scaling a scale may have two possible answers: YES/NO or AGREE/DISAGREE However, it is more common to find a four- or five-point Likert scale in operation. A Likert scale requires respondents to indicate a degree of agreement or disagreement with a statement or set of statements concerning a particular object. The Likert scale is popular because it is easy to construct and administer. General rules in question design 1. Avoid leading questions. 2. Avoid double questions 3. Avoid unfamiliar words or phrases 4. Vague and general questions should not be used 5. Hypothetical questions are of little value- they are difficult to implement and potentially unreliable General rules in question design 6. Use filter questions where necessary 7. The use of negatives in questions makes it difficult to understand 8. Long and overelaborate questions may not yield accurate answers 9. When asking questions on sensitive topics you should first consider whether you need the information 10. Questions which require the respondent to recall something that happened some time ago should be avoided if possible  Demographic Lecture 10 questions Demographic questions Most surveys include demographic questions in order to classify the respondents by age, life- stage, socio-economic group, gender and so on. These questions are now quite standard and examples are provided below. The researcher needs to consider whether such questions are required given the research aims, as they may lengthen the questionnaire unnecessarily. Piloting the survey In this way the whole procedure may be checked to determine whether or not things will run smoothly. Generally, a pilot will not be necessary if the survey has been well designed, but how do you know unless you pilot it first? The aim of the pilot survey is to test the reliability and validity of the survey. Response rates The success of the survey is measured to a large degree by its response rate. Activity 1: Draw up a list of advantages and disadvantages of face-to- face interviews and postal surveys. Activity 2: A poorly designed questionnaire The survey below is an example of a survey of customers of a leisure center designed by a student. The questionnaire is to be administered by a number of interviewers (fellow students) within the given set of neighborhoods. Identify and correct the errors.  Sampling Lecture 11 Techniques  Learning Outcomes At the end of this chapter you will be able to: 1. Formulate appropriate sampling strategies. 2. Distinguish between probability and non-probability samples 3. Use a variety of sampling techniques 4. Calculate the optimum sample size for probability samples Introduction The purpose of this lecture is to introduce a range of methods by which you can select people or things from which you plan to obtain data. Sampling techniques are closely associated with quantitative methods such as the survey and the experiment.  Sampling is simply a process of selecting participants for a piece of research. It is the means by which we obtain a sample, or a portion of the survey population. Thus, a sample is a sub-set of the population selected for inclusion in the research.  So, the first questions to ask when seeking to compile a sample are:  Is the population known and clearly defined (the population being defined by research aims)?  Can the population be listed as a sampling frame? If the answers are 'yes' to both these questions then sampling can proceed simply and systematically, using probability sampling.  A key outcome of most questionnaire surveys is the ability to generalise to the population-to say something meaningful about the situation in general rather than just restricted to the sample.  Sampling Lecture 12 strategies  There are two key questions which have to be addressed in any sample survey: 1. How should the sample be obtained? 2. How large should the sample be? The answers to these questions will be based on a consideration of sampling theory, the nature of population.  The first question can be broken down still further into a range of sub questions.  Firstly, who or what should be sampled?  Secondly, when should the sampling take place?  Thirdly, where should the survey be administered?  There are basically two generic types of sampling: probability sample every item in the sampling frame has an equal chance of being included in the sample. In other words, a probability sample is a technique which ensures a random sample,  non-probability sampling : Where a sampling frame does not exist, this kind of sampling is not usually possible and non-probability sampling is used.  There are several different types of non-probability sampling but all have one thing in common - not all elements have an equal chance of being selected. Such samples are not random and the degree of sampling error cannot be determined. Systematic sampling Systematic sampling may be preferable because the sampling frame is too large. This is one of the most direct and least expensive sampling methods.  Stratified sampling A stratified sample involves two stages: 1) The working population is divided into homogeneous sub-parts (strata), such as women and men, or large cities and small cities. 2) Random samples are taken from each sub-part  Summarising Data - Descriptive Lecture 13 Statistics and the Graphical Presentation of Data  Learning Outcomes At the end of this chapter you will be able to: understand the four 'levels of measurement. set up an SPSS data file. summarise data using tables and frequency distributions. use SPSS to generate frequency distributions. construct a range of graphical presentations to summarise data, and understand their appropriate use use SPSS to generate graphs. use SPSS to select cases for analysis. use SPSS to compute a new variable.  Data entry in SPSS Now that the variables have been set up, the data can be entered from the data file in Appendix 4. This is done in exactly the same way as you would enter data on any spread- sheet, using the arrow keys or the mouse to move around. At first the cursor will be in the first row of the first column. The frame of the cell is shown in bold to indicate that it is the active cell. Once all the data has been entered, check it for accuracy, and then save the data file in the usual way i.e. by selecting Save from the File menu. If you make any errors in entering values they can be corrected by highlighting the appropriate cell, entering the correct value, and moving to another cell. If you want to delete the contents of a cell and leave it empty, move to that cell and press the Backspace or Delete key and move to another cell. This will leave a full stop (.) in the cell to denote a missing value. To leave SPSS select Exit from the File menu.  Frequency distributions Question 3.a in the Hadrian's Wall survey recorded the length of holiday of the respondents. The answers of the first 100 respondents are shown in Table 9.3. There are, in fact, only 93 figures in Table 9.3 as seven respondents did not answer this question. As it stands, this is a fairly meaningless jumble of figures. The first step that we can take to clarify the picture is to construct a univariate frequency distribution where the various holiday lengths are listed along with their frequency in the distribution. This produces the following:  Simple bar charts Bar charts can be presented either horizontally or vertically. Simple bar charts are made up of a number of separate bars, and the height or length of the bar represents the size of the data. For example, the data generated by Question 13 in the Hadrian's Wall survey (Appendix 2) can be shown in the form of a bar chart (see Figure 9.4). This figure shows that most people got information on the area from a guidebook or from friends and family when planning their visit.  Pie charts An alternative way of showing the relative size/importance of groups is to use a pie chart. The data showing the socio-economic profile of visitors to Hadrian's Wall could also be presented in the form of a pie chart (see Figure 9.7). Here each segment or slice of the pie represents the percentage of cases falling in each socio- economic group. The SPSS Chart Editor makes it possible to show the percentage values and/or the actual values foreach segment of the pie. Pie charts are better than compound bar charts where there is a large number of categories but even here the ability to decipher information and to label and colour/shade the segments becomes more difficult as the number of segments increases. While there is no rule here, a pie chart containing more than six segments will become difficult to decipher. It may, therefore, be necessary to collapse categories for ease of presentation and under- standing, although in doing so you should ensure that important detail is not lost. Having said that you can have too many segments in a pie chart, you can also have too few. You should not, for example, produce a pie chart showing the proportion of respondents answering yes/no to a particular question. Charts should be used to summarise and clarify - in this case this can easily be achieved in a line of text. Neither should pie charts be used where time series analysis is involved, as it is difficult to compare a number of pie charts even if they are placed alongside each other Thank You