Contents:
1. THE RESEARCH PROCESS
2. QUALITATIVE AND QUANTITATIVE RESEARCH 3. NO LECTURE (WESTPAC SEMINAR — LIVE CASE) 4. EXPERIMENTAL RESEARCH & TEST MARKETING 5. PRELIMINARY ANALYSIS
6. WORKING WITH DATA USING SPSS 7. TESTS OF DIFFERENCES
8. TESTS OF ASSOCIATION 9. MULTIVARIATE ANALYSIS I 10. MULTIVARIATE ANALYSIS II
11. COMMUNICATING RESEARCH INSIGHTS
12. REVISION
Lecture 1
THE RESEARCH PROCESS MARKETING
- Customer’s point of view, collectively unique - Satisfying needs and wants
- Delivering service and satisfaction better and more effectively, needs change
Therefore, research is needed to generate the unknown insights, in-depth observation and robust, supported with evidence.
Why?
- Organisations are increasingly competitive with dynamic environments/markets.
- We are living in a disruptive/sharing era.
- Technological advancements - Evolving consumer values
- Staying in touch with their customers – attracting new customers
3 key conflicting demands from MR clients seeking faster and deeper insights: 1) accuracy and speed 2) leveraging technology/maintaining research integrity 3) robust/storytelling
The MR industry needs soft skills (interpersonal, story telling, contextualising), hard skills (data crunching, SPSS, Excel) and project management skills (discipline, detail, patience).
MARKETING RESEARCH
Systematic and objective process of generating information to aid marketing decisions
- Specifying information required to address market/social issues - Designing method for collecting information
- Managing and implementing the data collection process
- Analysing results
- Communicating the findings and their implications
Basic (pure) research: expand limits of knowledge, learn more, not aimed at problem solving, provide foundation and theory
Applied (engaged) research: decision must be made, aimed to understand/answer question USEFULNESS
- Is there sufficient time available before a managerial decision must be made?
- Are there sufficient funds to (i) conduct the research and (ii) implement the recommendations derived from the research?
- Is the existing information available adequate for making the decision? If not, can appropriate information be made available?
- Is the decision of considerable strategic or tactical importance?
- Does the value of the research information exceed the cost of conducting the research?
RESEARCH PROCESS
1. Defining research problem
Problem discovery as only symptoms of the problem may be apparent.
Finding the leakage is defined as the ‘problem definition stage’ à helps set proper research objectives allowing for a sense of orderly direction and investigation.
1 Defining the problem
2 Planning the research
design
3 Planning
the sample 4 Collecting
the data 5 Analysing the data
6 Formulating conclusions/
writing the final report
The normal distribution
A symmetrical, bell-shaped distribution that describes the expected probability distribution of many chance occurrences.
99% of its values are within ±3 standard deviations from its mean. Therefore, a standardised normal distribution, 1) symmetrical about its mean 2) infinite number of cases (area under the curve has a probability density equal to 1.0, 3) has a mean of 0 and standard deviation of 1 à
Must ask: do we keep outliers?
Z score
Also known as generating a standardised score for each raw score in a distribution.
Finding how many SD it is away and helps with identifying outliers. For scale data only. Z score = (X- M)/SD
Frequency distribution
A process of recording the number of times a particular value of a variable occurs.
Percentage distribution is a distribution of relative frequency. It is organised into a table that summarises percentage values associated with particular values of a variable.
Central limit theorem and confidence levels
States that as the sample size increases, the distribution of the mean of a random sample taken from practically any population approaches a normal distribution.
A confidence interval estimate is based on the knowledge that the population mean is the sample mean -/+ a small sampling error. This will help determine how probable it is that the population mean will fall within this range of statistical values
- Point estimate: an estimate of the population mean in the form of a single value, usually the sample mean
- Confidence level: a % that indicates the long-run probability that the results will be correct.
à The data analysis process can also involve testing hypotehses and/or establishing statistical significance.
HYPOTHESIS TESTING
An experiment of a hypothesis, where it is an empirically testable statement that is an unprove supposition (opinion) developed in order to explain a phenomena.
A null hypothesis: statement about a status quo
Alternative hypothesis: indicating the opposite of the null hypothesis
Deals with an implied relationship between (specific) variables to determine if differences or associations exist. The hypothesis testing procedure is: