Research

1
Research

• Chapter 5

2
PRs RACE Process

• Chapter 5, Research, begins four chapter sequence
  on the PR RACE processResearch, Action,
  Communication, Evaluation
• Before any PR program or campaign can be
  undertaken, information must be gathered and data
  must be collected and interpreted.
• There are different types and methods of
  researchdepends on the particular subject and
  situation.
• Chapter examines various research techniquesfrom
  using secondary sources such as libraries, books,
  newspapers and magazines, the Internet and
  online data bases and their respective strengths
  and weaknessto primary research such as
  questionnaires, interviews, sampling, and focus
  groups.

3
How PR Pros Use Research

• To achieve credibility with management
• Define audiences and to segment publics
• Formulate strategy
• Test messages
• Help management keep in touch
• Prevent crises
• Monitor the competition
• Sway public opinion
• Generate publicity
• Measure success

4
Casual and Informal Research

• In public relations, research is used to gather
  data and information
• Unlike scientific and other technical research,
  PR research is more casual and informal than
  scientific and precise, according to a survey of
  PR practitioners

5
Primary and Secondary Research

• Primary Research methods gather new and original
  information firsthand
• Methods include in-depth interviews, focus
  groups, surveys and polls
• Secondary Research uses existing information in
  books, magazines, newspapers, the Internet
• Also can involve archival research taking an
  inventory of organizational materials that can
  shed light on such things as
• The success of the product/service in the past
• Analysis of what geographical areas provide the
  most sales
• A profile of the typical customer who buys
  product/uses service

6
Rosetta Stone Example

• Rosetta Stone is a CD-Rom and online language
  learning software program
• PR campaign to increase brand awareness and sales
• Customer research showed
• More than 70-percent of purchasers were men,
  average age 38, highly educated and affluent
  (60,000 annual income)
• Sales were highest in the fourth quarter- just
  before Christmas with many customers receiving
  product as gifts
• Given this baseline data, the PR firm of
  Carmichael Lynch Spong was able to figure news
  strategies to reach out to new customers and core
  customers in other months of the year

7
Other Research Categories

• Qualitative
• Good For
• Probing attitudes/perceptions
• Assessing message penetration
• Testing messages
• Qualitative Methods
• Content Analysis- gather clips
• Interviews- intercept/mall interviews
• Focus Groups-8-12 people
• Copy Testing- is this clear?
• Ethnographic Techniques- observing
  patterns/behavior roll playing

• Quantitative
• Characteristics
• More hard than soft data
• More scientific
• Highly structured
• Needs random sampling and large number of
  respondents
• Quantitative Methods
• Telephone polls
• Mail surveys
• Shared cost studies

8
National Survey Surprises?

• To get a highly accurate idea of what the U.S.
  adult population is thinking, national polling
  firms usually sample 1,000 to 1,500 people
• Effective polls and surveys require a random
  sample (aka probability sample)
• This means that everyone in the targeted audience
  has an equal or known chance for being selected
  for the survey
• The average national poll samples 1,500 people
  and the margin for error is within 3 percentage
  points 95 percent of the time

• This means 19 out of 20 times the same
  questionnaire is administered, the results should
  be within the same 3 percentage points and
  accurately reflect the whole population
• Telephone directories are often used to randomly
  select respondents
• Hawaii example- p.140
• PR sample sizes are usually smallercan
  accurately get a sense of public attitudes and
  opinions from 250-500 responses gives plus/minus
  5-6 percent accuracy (see p.140)

9
Questionnaire Guidelines

• Much care must be taken in the wording of all
  questions to ensure maximum clarity
• Questions often go through many drafts before
  achieving the desired tone, clearness
• Avoid biased, slanted wording
• and questions Example (from an ACLU
  survey)Do you agree with this statement I
  believe that the President does not need to use
  unauthorized and illegal powers to keep us safe,
  that warrantless spying on Americans is
  unnecessary, and illegal and that, in America, no
  one including the Presidentis above the law.

10
Answer Categories

• It is important that answer choices are provided
  that cover a range of opinions
• Avoid yes or no responses they dont really
  provide much insight into perceptions and
  attitudes
• Seek instead to have answers that probe the depth
  of feeling among respondents and may serve as
  guidelines for management in making major changes
  or just fine-tuning existing policy

11
Specific Questionnaire Guidelines

• What type of info needed and in what detail
• State survey objectives in writing
• Decide who/which groups will receive survey
• Decide the sample size
• State the purpose of the survey
• Guarantee anonymity
• Use closed-end (multiple choice) answers as often
  as possible
• Design in way that answers can easily be coded
  for analysis

• No more than 25 questions
• Use ranges for age, education, incomepeople more
  comfortable
• Use simple, familiar words
• Avoid ambiguous words and phrases that may
  confuse
• Context and placement of questions are important
• Provide space for comments at end
• Pretest the questions with reps. of your target
  audience for understanding and possible bias

12
Reaching Respondents

• The Pros and Cons of
• Mail Questionnaires- Easy to distribute Low
  response rate
• Telephone Surveys- Feedback immediate, personal,
  skilled phone interviews can get high response
  rate Hard to get hold of phone numbers
  skeptical public Trying to sell me something?
• Personal Interviews- Can generate a wealth of
  detailed information Expensive, smaller response
  pool, much advance work
• Omnibus or Piggyback Surveys- can buy a few
  questions that will be part of a national survey.
  Plus is cost. Minus is may only get a small
  snapshot of public opinion with one/two questions
• Web and E-mail Surveys- Benefits are that large
  samples can be generated in a short amount of
  time, economical, data can be analyzed
  continually. Negatives are No control over
  sample size or the selection of respondents.
  Attract respondents with banner ads, e-mail,
  phone, postcard invitations, and offering gift
  coupons and other incentives to respondents