Surveys. A survey is a set of carefully planned questions used to gather data with a particular objective or goal in mind These questions can be completed.

Surveys

• A survey is a set of carefully planned questions used to gather data with a particular objective or goal in mind

• These questions can be completed in writing or orally, in person, on the phone, through the mail, or on the internet.

• the main purpose of a survey is the fact that it is impossible to question everyone in a “population” or targeted area.

• It is impractical (too much time, money and effort)

• As a result, we use a survey on a selection of our target market to get sample responses that we can confidently project onto the greater population as a whole

• In order to confidently be able to project our findings into our marketing efforts on the greater “population”, we need to ensure the survey is properly created and conducted

An effective survey will consider each of the followingelements:

1. Define the goal of the survey

2. Design good questions

3. Determine how it will be conducted

4. Analyze the results

5. Take effective action

1. Define the goal of the survey

For a survey to fulfill its function it requires a stated goal that it is working towards.

i.e. Should we change the shape of our product?

Should we offer our product in different colours?

Who buys our product?

What added features would customers like?

2. Design good questions

1. Write questions that are simple and to the point2. Use words with clear meanings 3. Limit the number of ranking options 4. In a multiple choice question, cover all the options

without overlapping5. Avoid double-barreled questions 6. Offer an "out" for questions that don't apply 7. Avoid offering too few or too many options 8. Make recall easy

Types of questions

• A critical part of any survey• The questions must be chosen and constructed with great strategy• There are two main types of questions:

• Closed Ended

• Open Ended

Closed Ended Questions

• Most surveys use closed-ended questions, which ask you to select an answer from two or more choices.

• These questions “funnel” the respondent’s answers to a set of provided responses

• i.e. – yes or no, male or female, I like or dislike

• they are quick and simple to answer, easy to sort and analyze

Here are some sample closed-ended questions:

• I would buy this brand again

agree uncertain disagree

• Are you a smoker?

yes no

• Which brand names have you purchased in the last year?

Dove Ivory Zest

Lever 2000 Irish Spring Other

• Rate our service using the following scale: poor

excellent• Clean 1 2 3 4 5• Friendly staff 1 2 3 4 5

• What do you look for when you shop for a DVD player?

simple remote ease of use plays MP3/JPG brand name multi-disc warranty

Open Ended Questions

• Open ended questions allow the respondent to answer the question how they see fit

• In many cases, the researcher doesn’t know what type of response they will receive

• The information received is often very rich and detailed

“What is your opinion of Air Canada?”

“When I choose an airline, the most important consideration in my decision is …

“I flew Air Canada a few days ago. They gave me a cold sandwich toeat. This stirred the following thoughts and feelings in me...” Now complete the story.

3. Determine how it will be conducted

• Sample

• Sample Size

• Randomness

• Bias

Sample

• Instead of polling each person, a "sample" is scientifically selected so that results are representative of the whole group

• a sample must be created so that it includes a “representative” picture of the greater population

• i.e. – surveying a Grade 9 boys gym class about high school students’ attitudes towards the cafeteria….is this representative?

Samples are rarely ever 100% accurate, but the bigger the sample, the more likely it is valid

• Increased size reduces risk-but are also more time consuming and expensive

• The Canadian Census is considered a "100% Sample" because every Canadian is accounted for

i.e. - To determine the attitudes of GoodLife Fitness members about new start times

• it would be a valid “sample size” if 100 members were sampled in a population of 2,000

• It wouldn’t be a valid “sample size” if 3 members were sampled in a population of 2,000

Sample Size

Randomness

• A sample is only an accurate representation of the whole if all parts of the whole have an equal chance of being sampled – this is what is referred to as “random”

• If the participant’s are chosen on any other basis than “being random”, it could lead to skewed results

i.e – To determine the attitudes of GoodLife Fitness members about start times.

• it would be “random” if 100 members were chosen randomly from a bin with all 2000 members’ names inside

• It wouldn’t be “random” if a trainer picked the members that came in between 1:00 PM and 2:00 PM

Bias

• A survey is biased if it tends to lead the respondent to select a particular opinion

• As marketers, we want to know what the respondent is thinking and not influence their response

• i.e. - To determine the attitudes of GoodLife Fitness members on new start times

• it would be “biased” if you asked “isn’t insane to keep the gym open later?”

• It wouldn’t be “biased” if I asked “what time would you want the gym to stay open until?”

4. Analyze the results

• Tabulate the raw data

• Determine the results

• Plot a course of action based on results

5. Take effective action

Implement course of action