Marketing Research: Definition and Tasks

• Marketing research involves studying decision problems, research questions, and types of data.

What is Marketing Research?

• Gathers information about consumers' needs and preferences.
• Helps secure a competitive edge.
• Crucial for identifying and analyzing market needs, size, and competition.
• Most effective when addressing a problem.

Definition

• Connects an organization to its markets by generating information.
• Helps to find the best solutions to decision problems.

Tasks

• Defining the scope of marketing research.
• Designing the research approach.
• Implementing the research plan.
• Communicating the research results.

Decision problems and research

• Management faces a situation where they need to decide on an action to achieve a specific goal.
• Marketing research generates understandable, accurate, relevant, and timely information for problem-solving.
• Decision problems arise from weaknesses, threats, strengths, or opportunities.

Decision Problem instances

• Strengths: Capabilities like attracting clients with high willingness to pay for quality.
• Weaknesses: Ineffective marketing or wrong product mixes.
• Opportunities: Competitor leaving the market.
• Threats: Changing laws or increased competition.
• Market research considers unexpected challenges, future opportunities, and growth.
• Discovery: A new idea, event, or product.
• Evoke: Trends, insights and opportunities.
• Provoke: Problems and challenges.

Marketing research phases

• Determine the research scope.
• Select the research method.
• Gather and prepare the data.
• Analyze the data.
• Transform the results into actionable information.
• Data analysis provides usable information for decision-makers.

Types of Data

• Primary or secondary data.
• Primary includes words, numbers, or observations for research.
• Information is structured for managerial reports.
• Information is managerial interpretations and story telling.

Secondary data

• Data that was previously collected by someone else for another purpose.
• Government sources are quantitative and provide official statistics.
• Qualitative sources are very numerous including government reports and newspapers.
• Primary data is collected firsthand by the researcher.
• Researchers use questions and methods to get the data and generate primary data.

Internal vs External Secondary data

• Internal data is accessed and saved quickly, so is from the company.
• Internal data might not be up to date or relevant.
• Budgets and sales reports, call reports, inquiries, shipping records, distributor feedback, cost and marketing data are sources of secondary data.

Data Mining

• A process of discovering patterns in large datasets with methods in machine learning, statistics, and databases.
• Process: collect, clean, store, search for pattern, review, find, review/refine and action.
• Requires a data warehouse mapping scheme.

Data Mining Activities

• Predictive: classification, regression, time series analysis, prediction.
• Descriptive: clustering, summarization, association rules, sequence discovery, visualisation.
• Verification: goodness of fit, hypothesis testing, ANOVA.

Issues in data mining:

• Integrating new knowledge with existing knowledge.
• Handling noisy or incomplete data.
• Dealing with distributed, complex, and performing data effectively.
• Protecting data privacy and security.

Process and Problems with Data Mining

• Data mining: Lack of attributes.
• Data cleaning: Filtering to cover valuable information.
• Data integration: Null values due to connecting tables.
• Software Implement: Mistakes during data input.
• Software Designing: coding disaccord between systems.
• Company planning: Lack of data mining plans and lack of functions.

Ethics and Legal Concerns

• Combining info could violate privacy laws.
• Must tell people how their data will be used.
• Ethnicity should not be a factor in loan approvals.
• Job applicants cannot be screened based on age or sex.
• Insurance coverage cannot be declined based on neighbourhood.
• It may be ethical to consider ethnicity for medical diagnoses and treatments.
• Job loss can occur due to automated decision-making and can be inaccurate due to DSS.
• Data ownership and security are key.

Data Mining

• May create profiles with discrimination, privacy, and security.
• Need to be carful when singling out people for searching due to age, residence and race.

Types of secondary data

• Internal and external data for commercial.
• Indexes guides, indexes etc.

Data Comparison

• Primary: Data is collected directly by the researcher with no need for adjustments.
• Secondary: Data is collected by others.
• Primary data sources: Surveys, observations, and experiments.
• Secondary data sources: Internal records, government publications.
• Important to identify the target population.

Primary data

• New data collected directly.
• Secondary data: Public/existing data are collected by others.
• Interviews can get in-depth info via samples.
• Surveys and questionnaires have an adquate time to respond, so they’re cheaper.
• Social media monitoring is quick and real time.

Observation pros and cons

• Pros: Data is objective, unaffected by past/future.
• Cons: Limited info, expensive.
• Focus groups: Low cost interviews stimulate ideas, so responses may be bias.

Research complexity

• Increased opportunity and challenges from technology.
• Lots of choices in tools makes it tough to choose a right method.
• Companies are choosing hybrid research methods for various methods.

Surveys

• Creating a survey with Qualtrics.
• Agree or withdraw from participation with survey options.
• Use quality perceptions.
• Indicates a level of agreement that each gives high quality.
• Single and matrix options for mobile friendly tick boxes.
• Indicate products that are valuable for money.
• Write five statements or less.
• Measure overall attitude to a brand against competitors for brand equity.

Working with Qualtrics and Excel/SPSS

• Export data to word.
• Use numeric values in data and export.
• Recode seen but unanswered questions -99.
• Make data readable by starting at 1, not test.

Data view

• Missing data = seen, no response is -99.
• Data is not seen by respondent.

Data Statistics

• Quickly create descriptive statistics and charts.
• Right click names for variables.

Real Data

• Must remove incompletes and invariants.
• Remove missing data.
• Add coding.

Adjust data

• Cleaning and removing data is important.
• Fix all data sets so it makes sense.
• Run statistics after fixed data.

Data Manipulation

• After data import, use SPSS.
• Transform for new variables.
• Create a case for each respondent.

Recoding variables

• Reverse coding in variables.
• Compute variables.

Creating New Variables

• Use scales.
• Collapse cells.

Recoding scales

• To numeric.
• Group data together.

Quantitative Data

• Analyze, study and create reports.

Analytical Applications

• Compare 2 group means.
• Specify/ test the hypothesis.
• Preform/interpret the output.

Categorical Variables

• Run coupon samples.
• Run test by looking and seeing the means equal to 0, and the means that are unequal to each other.

Statistical significance

• P value should be close to 0.05.
• Three forms of hypothesis tests, upper, lower and 2 tailed.
• Nulls use is equal to, alternative means use is significantly.

T-tests

• Used to know what there is significant difference in means.
• Need to see if males and females differing in average spending.

SPSS testing

• See if there is a significant difference in customers spending.
• Graph the histogram and variable amounts.
• Used the Levine test for quality with equal variances.

Test for One-Tailed Data

• Divide the data by two to give each significance more importance.
• Then data is useful.