Analysis of the Data

Analysis of the Data

1. Exploratory Data Analysis

   • Coding Open Ended Questions
     • Open ended questions can be used in postal questionnaires as well as in interviews. The usual reason for using open-ended questions is that the researcher has no clear hypotheses regarding answers, which will be numerous and varied.
       • At the minute I have no clear hypothesis regarding answers, and therefore coding open ending questions is very suitable to my research.
     • The best way to analyse open-ended questions is to code the information in terms of ideas and themes.
       • I should use the example found in the book to analyses any open-ended data that I create. This can be found on pages 99-103 in Naoum (1998)
   • This type of data analysis refers to research about peoples lives, their stories and behaviour, and it can also be used to examine organisation, relationships and social movements. Some data may be quantified afterwards, but the analysis is qualitative.
     • I think my research is more of a cultural movement and is based a lot around how people will take to a new technology. Meaning my research will suit this type of analysis.
   • Research carried out in this way produces descriptive data such as people’s own spoken or written words or observable behaviour.
     • As my research is going to be qualititave and will involved lots of peoples own spoken or written words, coding open-ended questions will need to be applied.

2. The descriptive statistics

   • Mesaure of central tendency
     • This type of analysis is applied when you have a group of data and you wish to find the most typical value for the group, or the score which all other scores are evenly clustered around. These statistics are known as the ‘mean’, the ‘median’ and the ‘mode’.
       • Mesure of central tendancy can be used to find establish the most typical value for the group of respondents
   • Normal distribution
     • Normal distribution is an important expression in the field of statistics because the selection of some inferential statistics depends on whether the data is normally distributed or not.
   • Frequency distribution
     • When summarising large amount of raw data it is often useful to distribute the data into categories or classes and to determine the number of individuals or case belonging to each category. This is called ‘category frequency’.
       • Tabulation
       • Bar Charts / Histograms
       • Pie Charts
       • This could be used to analyse the interview or questionnaire responses that I will receive with my research. Bar charts could be use to demonstrate the amount of people that think that they would use automated procurement in their organisation, for example.
   • Standard Deviation
     • The standard deviation is another measure of the degree in which the data is spread around the mean. The theoretical basis of the standard deviation is complex and need not trouble the original user (most scientific calculators have a function to calculate the standard deviation).
       • Standard deviation and normal distribution may be to complex and unnecessary for the data I am going to produce. This meaning I shall not use it to analyse my data.

3. Steps to be carried out before and after the inferential statistics

   • Step 1 – Formulate your research hypothesis in terms of the predicted results
     • Step 2 – State the Null hypothesis
       • Step 3 – Decide which test to use
         • Step 4 – Calculate and obtain the test statistics
           • Step 5 – Decide whether result is significant (depends on significant level set)
             • One these steps are complete then can move on to The inferential statistics.
   • At this moment in time I do not think that my research would entail enough analysis of numbers to require inferential statistics

4. The inferential statistics (bivariate)

   • Parametric test (The t-test)
     • The parametric test which is used to compare the difference between the mean score two samples.
   • Non- Parametric test (the chi-square test)
     • The chi-square is a test of association between two set of data and can be used only when the data is nominal or ordinal. It predicts how many subjects in each group will fall into certain categories.
   • Spearman rank correlation coefficient
     • The spearman (rho) correlation is a non-parametric test for measuring the difference in ranking between two groups of respondent’s scoring a number of issues, attributes or factors.
   • Pearson product moment correlation coefficient
     • The correlation coefficient is a parametric test which is used when you have two sets of score and you want to calculate whether there is a strong relationship between them.
   • I do not think that my research would entail enough analysis of numbers to require inferential statistics

5. Writing highlighted in Green represents information and headings taken from the textbook.

   • Writing highlighted in blue represents useful quotes taken from the literature.
     • Writing highlighted in Orange represents my comments and thoughts.
       • Writing highlighted in Purple represents analysis methods that will be used in my research.