exam 2 Flashcards
4 essential steps of public relations
Research
Planning
Communication
Measurement
research is defined as the
systematic investigation
of a problem
involving gathering evidences
to make inferences
systematic: predefined procedures and methods
survey questionnaire
guidelines for focus group
codebook for content analysis
systematic: inter subjectivity
It must be possible for other researchers to replicate our study and come to the same results.
Of a Research Problem…
Should be empirical, should be answered based on observable evidence
3 key types of problems:
Exploratory- (focus groups to understand voters’ reaction to new policies)
Descriptive-(audience research, research of market shares)
Causal-(examining the influence of one variable on another)
exploratory research
Pretesting brand and product names in different cultures.
Often, up to 10,000 different variations tested
exploratory research involves a focus group for
Cognitive associations: sports → football
Different meaning (a word can have two separate meanings in different languages) (Ford Probe translates into Ford Trial in German)
Pronunciation (if it’s hard to pronounce, less people will buy it)
When gathering evidence, search for social regularities
Predictions about specific publics, larger group of voters, or consumers
NOT predictions about individuals.
when gathering evidence, there’s always a chance of error
finding hold within some margin of error, there’s never absolute certainty
important distinction: probabilistic vs. deterministic predictions - we can only predict how people will most likely react, without certainty
Most market and PR research is still based on samples. The goal to make inferences to:
A larger population
Other time periods
Other locations/societies
academic research
often called “basic” research
funded through universities or foundations in order to answer broader theoretical questions
conducted by academics
data remain property of the researcher but usually can be used by other researchers.
applied research
often called “industry” research funded by corporate or political sponsors to answer a specific, applied question conducted by -academics -research departments of larger firms -market research or consulting companies data remains property of the client
primary research
Information gathered by the researchers through person-to-person interaction.
Can be gathered through meetings. one-on-one interviews, focus group, surveys
secondary research
Information gathers through available literature publications, broadcast media, and other non-human sources. Generally easier to gather than primary (Nielsen data, Pew data,)
quantitative research
“numerical tabulations and statistical comparisons made possible by systematic surveys, experiments, observations, or analysis of records. Data are used to test hypotheses and identify the strength of patterns observed using qualitative methods”
qualitative research
“descriptions of cultural situations obtained from interviewing , participant observation, and collection of oral and textual materials”
popular quantitative methodologies
Surveys (telephone; mailed; online)
Sampling: random, area probability, snowball, convenience
Complexity, length of questionnaire
Survey mode (in person, telephone, email, web)
Analysis
when do we use qualitative methods?
When you’re in new territory and little is known
When customer perceptions or attitudes may be hidden from easy view
When the product category may represent unspoken meaning to buyers
To generate ideas for products, advertising, or brand positioning
To feed a formal idea generation process
To screen ideas and concepts.
qualitative data provides….
…insights into how and why people think and behave as they do
the most popular qualitative methods
Interviews
Participant - observation, and
Focus groups
structured interview
Uses an interview schedule and adheres fairly strictly to it
Similar to a survey in that the informants don’t really guide the interviewer
semi-structured interview
Begins with a key set of questions for interview
But, allows informats to wander into interesting territory
Later informants may be asked about these issues
unstructured interview
Freedom
One general opening statement and then a free flow.
focus groups
Given high costs of interviews, researchers increasingly turning to focus groups
Consist of 5-10 people who are chosen based on their relevance to the study
It is a guided discussion designed to explore a topic of special interest to the client/research
as a participant- observation
methodology- With the embedded method, you may choose not to record when you are interacting with others, but you remember and record your observations.
Your data are your observations
Your analysis involves comparing your observations to your earlier expectations.
cross-sectional research
Research based on a sample drawn at a single point in time
longitudinal research
Research based on one or multiple sample with measurements taken at taken at multiple points time
types of longitudinal studies
trend studies: data trends as they develop over time
panel studies: looking at same individuals over time
cohort studies: looking at how certain groups data collections over time
Probability Sampling: Each element of the population has a
non-zero (the chance of an element being selected is greater than zero)
known, (we know what that chance is, i.e. if there are 5 elements, each element has a ⅕ chance of being selected)
equal chance of being selected into the sample (again, if there are 5 elements, there must be a ⅕ chance for each)
non-probability sampling
One of the assumptions of probability sampling is violated, T.t call-in polls, Internet surveys (if you don’t have TV or internet, it is ineffective)
Important: In order to calculate sampling err (measure of how precise a poll really is) we need to use probability sampling techniques
probability- simple random sample SRS
Every element and every combination of elements has an equal chance of being selected Problem: Requires a list of sampling units (ex: a class list)
probability- systematic sampling
Choose a starting point and every nth unit is chosen to be selected
Problem: Requires a list of sampling units (ex: “ “)
probabilit- cluster sampling
Divide population into clusters and then randomly choose X number of clusters.
non-probability sampling methodologies
Accidental/ Haphazard/Convenience Sampling
Exit Polls
Call-in Polls
purposive/ judgmental sampling
“Typical” precincts and medical research (going to take awhile)
targeting people non-probability based
snowball sampling
Invitation based on your sample
For example, if you make a facebook event and invite a certain number of people, set it so that they then can invite whoever they want, and so on, it “snowballs”
tend to be biased if we all connect with the same people
quota sampling
Once most common sampling technique
Involves sampling according to predefined characteristics of population (race, gender, sex, etc…)
sometimes too much freedom for researcher
why do we sample? why not just ask everyone’s opinion?
Not feasible to get every persons opinion
small control group to apply to everyone
sampling allows their opinion is what everyone else is saying (with room for error)
The Development of Opinion & Market Research
Before 1940s
Naive outlook toward research based on
WWI Propaganda
Orson Welles’ “War of the Worlds” (Oct., 1938)
Lack of commonly-accepted research procedures
“Magic Bullet”
the development of opinion and market research: During and immediately after WW II:
Involvement of government in applied research
Columbia’s “Bureau of Applied Social Research”
Lazarsfeld et al.: The People Choice
New developments in research techniques
“Reinforcement and two-step flow”
the development of opinion and market research: during and after 1950s
Sharp increase in privately-funded PR and advertising research
Emergence of telephone and online polling
As a result: development of (industry/ research as we know it today)
“Contingent and delayed effects”
ways2useresearch Achieve credibility with management
executives want facts, not guesses. The inclusion of PR personnel in an organization’s policy and decision making is strongly correlated with their ability to do research and relate their findings to the organization’s objectives.
ways2useresearch define/segments policy
information on demographics, characteristics, lifestyles and consumption patterns of audiences helps ensure messages reach the proper audiences
ways2useresearch: formulate strategy
You want the right strategy because otherwise, you may waste money
ways2useresearch: test messages
Determines what messages work best for the target audience.
ways2useresearch: prevent crises
Uncovers trouble spots and public concerns before they become page one news. This tips off an organization that they
ways2useresearch: monitor competition
i.e. surveys that ask consumers to comment on competing products. This helps an organization shape its marketing and communication strategies to position a product and capitalize on competitor’s weaknesses.
