MRG Tools of the Trade Introduction to quant and other data types and Big Data John Carroll, 26 February 2018 @MediaCarroll 1
What you ll take away Understanding of what methods and data types are available when and where used the advantages and limitations current industry challenges 2
Introduction to the key data types Qualitative Research is primarily exploratory research. It is used to gain an understanding of underlying reasons, opinions, and motivations. Quantitative Research is used to quantify the problem by way of generating numerical data or data that can be transformed into usable statistics. Big Data is a term that refers to extremely large data sets that may be analysed computationally to reveal patterns, trends, and associations, especially relating to human behaviour and interactions. 3
Share of spend by research method Online 27% Total Qualitative 15% Other 14% Total Quantitative 71% Automated digital/electronic Telephone Face-to-face Online traffic/audience Mobile/Smartphone 12% 8% 7% 6% 5% Mail 1% Other quantitative 4% Group discussions 6% Online communities 6% In-depth face-to-face 2% Ethnography/blogging 1% Other qualitative 1% Other 14% Source: ESOMAR Global Market Research 2017 4
Key characteristics Need for representative samples Quant research Measuring attitudes, opinion, behaviour, knowledge, characteristics Statistically robust data Substantial number of interviews Structured questionnaires / closed questions 5
6
Interviewer administered surveys 7
Face-to-face interviewing Advantages Good response rates More difficult to refuse to a person in front of you Good rapport with participants Can show stimulus material Good for clustered respondent types, e.g. urban areas, some ethnic minorities, etc 8
Face-to-face interviewing Limitations Speed Difficult to monitor/ control Biases related to availability of participants/ ease of access for interviewer Reactive biases Cost Often limited to general public 9
Telephone research Advantages Can be set up and completed quickly Can brief centrally interviewers Control over quotas, interviewers, who responds Can listen in to/record interviews Complex routing Good for businesses, geographically spread audiences, international work Cheaper than F2F Most people have landlines 10
Telephone research Limitations Where sample coming from? Needs tracing, de-duping, ex-directory, formatting etc Less rapport with interviewer Constraints on interview length Questions/code lists need to be concise No visual stimulus (although possible) Easier to refuse over the phone Call screening Increase in mobile-only households Threats from direct marketing More expensive than postal and online 11
Self completion surveys 12
Postal research Advantages Good for large, geographically dispersed audiences Relatively straightforward to set up Cost effective No interviewer bias Ideal for sensitive questions Basic visual stimulus Can be completed in own time Participants have time to consider answers, collate information Potential use for online survey recruitment 13
Postal research Limitations Typically lower response rates Takes time Very reliant on Royal Mail and the printers No control over quotas, who responds (particularly if no named sample) Constrained over questionnaire length No complex routing More errors in data participants don t follow routing, instructions Opportunity for participants to discuss answers Open ended answers have to be taken off the paper questionnaire Potential data capture errors 14
Online research Advantages Can be cheaper than other methods (particularly with large sample sizes) Quick to set up and complete Can show multimedia stimulus No interviewer bias Ideal for sensitive subjects Complex routing and high quality data Can be completed in own time Verbatim answers easy to export Good for business/ international projects Easy/cheap to send reminders Real time reporting Online part of everyday lives for many Good for accessibility 15
Online research Limitations Where sample from? Not everyone is online issues of representation Possible participant technical issues Cannot guarantee response rate Some participants concerned about data security Flat-lining participants Possible researcher issues Who is responding? 16
17
Other quantitative techniques Auditorium Hall tests Self-completion sessions e.g. in school Omnibus Surveys SMS texting Pop-up surveys In-store kiosks In-store / entry / exit interviews Electronic diaries 18
Passive methods An image conveying passive Big brother? 19
20
How many types of quant research? 2 21
Good Bad 22
What are your objectives? Who do you want to speak to? How can you sample them? Where are they? How important is representativeness? How long is the interview? Do you need visual/audio stimulus? What subject(s) will you cover? 23
How quickly do you need results? What is the budget? 24
Challenges Quality Speed Cost 25
Hybrid 26
27
Big Data 28
29
30
What are your objectives? Who do you want data from? How can you identify them? Where are they? How important is representativeness? Can I integrate different data sets? 31
WE CANNOT SOLVE PROBLEMS BY USING THE SAME KIND OF THINKING WE USED WHEN WE CREATED THEM Albert Einstein (1879-1955) 32
Design matters 33
NEVER ACCEPT DATA at FACE VALUE 34
If it LOOKS INTERESTING then it s PROBABLY WRONG 35
Thank you For more information please contact john.carroll@gfk.com @MediaCarroll 36