Send the link below via email or IMCopy
Present to your audienceStart remote presentation
- Invited audience members will follow you as you navigate and present
- People invited to a presentation do not need a Prezi account
- This link expires 10 minutes after you close the presentation
- A maximum of 30 users can follow your presentation
- Learn more about this feature in our knowledge base article
Do you really want to delete this prezi?
Neither you, nor the coeditors you shared it with will be able to recover it again.
Make your likes visible on Facebook?
You can change this under Settings & Account at any time.
Transcript of Questionnaire Format
The 4 M's
The Problem Statement
An in-depth look at your Problem Definition reveals that it focuses on one of these four "M's":
1. The Market - Who is our consumer?
Who is the intended purchaser?
How many are there?
What are their characteristics, behaviors?
2. The Media - How do we reach them?
What media is most effetive and efficient?
What are the strenghts and weaknesses of each medium?
3. The Money - How much do we spend?
Do I stretch it out over the year or spend it in a quick blitz?
4. The Message - What do we say?
How do we say it?
What will trigger the sale?
Paley Library database
What do we want to find out?
Who should we have respond to our survey?
What is our sample size and confidence level?
When do we need the results?
What is our budget?
Develop three sample questions
Prioritize seven key factors
Numeric - percents, frequencies, averages
Examine through statistical techniques
Representative of the population as a whole
Fixed response options
Easier to analyze and interpret
There are 3 Types of Quantitative Research Techniques:
1. Observational - Collecting info through the recording of objects, events, situations, behaviors. Can be natural or in lab setting.
Simple and inexpensive
First hand information
Deeper understanding that surveys can't obtain
Observation effect may bias findings
Reliance on interpretation
2. Physiological - measures voluntary and involuntary response to stimuli. Shows observable behaviors but not the "why?" Includes:
Galvonic Skin Response
Brain Wave Analysis
3. Survey Research
high refusal rates
No personal interaction
Loss of control
Best for longer surveys
Convenient for respondant
highest level of credibility
works when Q's are long or complex
difficult to get true random sample
Researching the Consumer
Diaries and journals
Researching and Evaluating the Media
Syndicated secondary research
Media and vehicle effectiveness
the Media mix
Optional Timing, ad units
Good Questions are:
Avoid abbreviations and jargon
Asks one question at a time
Justifies the use of personal info
Doesn't make assumptions
Includes clear and logical responses
Introduction and instructions
3. Info needed and steps involved:
Bridges the gap between the problem and real world data collection
Ask yourself, "what do I mean by...?" What exactly do you want answered?
What ways can this be answered?
How will this be collected?
2. Justification of the research:
The value is greater than the cost to retrieve it
If research is not done, end result can be devastating
1. Describes (defines) the problem or opportunity:
Selecting an alternative
Explains a problem or opportunity
Broadens your knowledge of the consumer, your product or the market
Ease of Access
Low cost to acquire
May help to clarify research question
May answer research question
May show difficulties to expect in conducting primary research
Quality of researcher
Not specific to your needs
Not timely; Obsolete
When secondary research results are incomplete, it's time to do primary research
The first step in Primary Research is determining who you want to survey.
Census = everyone in the target population (can be huge, expensive and time-consuming). There is no statistical variance when using a census.
Sample = a subset of the target popultion. Useful for forming generalizations about the population as a whole, without surveying the entire group.
So when is it appropriate to conduct a census?
When the population is small and identifiable
When sampling might eliminate important points of view
When credibility requires that the entire population be included (for example, when you need detailed info regarding small segments of your universe)
Once you decide that you want to do a sample, you must decide on whether you want to do a probability sample or a non-probability sample
Probability Sample is random and scientifically chosen. Each member of the population has an equal opportunity to be selected, which creates an accurate reflection of the true characteristics of the popluation.
simple random sample - think of this as a drawing where every person in the population puts their name in a hat and has an equal chance of being drawn.
systematic random - you'll choose every nth listing.
stratified random - coincides with the demographics of your population.
Non-Probability sampling is unscientific and unsystematic. Includes:
Convenience sampling (accessible and easy)
Judgement or quota sampling (generalization and assumption)
Non probability sampling cannot be used to infer the results will reflect the population as a whole.
Despite their unreliability, they are frequently used in advertising and marketing because they are cheaper and take less knowledge, time, thought and analysis.
Good layout and visual appeal
uncluttered and easy to read
questions should be distinguished from responses
Coding should be unobtrusive
Questions should not continue across pages
Use columns to maintain focus, save space and simplify responses
Introduce the project
request for participation
Important and legit (the need for truthful answers)
Can quit at any time
Length of survey/interview
How/where to return survey
Give clear directions
keeps out those you don't want
allows those you want to continue
The main series of questions
First questions should be easy
avoid open ended questions in beginning
Address important topics early
Sensitive or difficult questions at the end
Arrange questions so that they flow naturally
Group questions of the same topics together
Complete one topic before moving on to another
Start with general questions then move to the more specific
avoid biasing questions appearing later in the questionnaire
Pre code when possible
useful for checking patterns or for cross tabs
7 Key factors:
The first step in defining the problem is determining the "type" of problem you have:
Are you selecting or evaluating alternatives?
Are you trying to determine if there is a problem or an opportunity?
Are you attempting to understand your consumer or the environment of your marketplace?
research is necessary when:
the value of the research is greater than the cost to acquire it
the negative consequences of making the wrong decision are high
Specify informational needs:
Evaluate courses of action
Identify info that responds to problem definition
Get your client's approval
Kinds of Information:
What people know
how well they understand something
Beliefs, Attitudes, Opinions
Behavior - what people do
what people are
what people have
Nominal Level Questions are used when the goal is only classification.
1. Mutually Exclusive
3. Simple to code and analyze
Dichotomous (1 of 2 groups)
Multiple Choice (with one response)
Checklist (multiple responses)
4. Assumes category equivalence
Ordinal Level informs the researcher of the ordering of items, typically from greater to smaller or from more to less.
As opposed to numbers on the nominal level, ordinal level numbers have some mathematical meaning.
HOWEVER, Does not provide insight into relative distance between ranked objects.
Interval Level provides an estimate of the relative distance between items
Ratio Level measures on a continuum which reflects the degree represented
sample size = (confidence level / confidence interval) squared x standard deviation x 1- standard deviation