Information
AI Chat

Study Guide Exam 2

study guide for exam 2

Course

Principles of Marketing Research (MKTG 352)

18 Documents

Students shared 18 documents in this course

University

University of South Carolina

Academic year: 2019/2020

Uploaded by:

Anonymous Student

This document has been uploaded by a student, just like you, who decided to remain anonymous.

University of South Carolina

Comments

Please sign in or register to post comments.

Preview text

Measurement Levels and Measurement Formats

Constructs

Ex. boba example from their conceptual model which includes Perceived price, age, gender, sweetness, product diversity, store ambience

● Meaure willingness to buy ○ I prefer to buy ○ I would return to buy ○ I would like to buy ___ ○ I like to buy ____

● Measuring perceived store ambience ● Measuring perceived product diversity ● Measuring perceived sweetness ● Measuring perceived price

Variable- Measuring Age and Gender - “What is your age” - Select your gender

Validity and reliability of constructs

● Validity is whether we are measuring the right thing ● Reliability is whether there is consistency is our measurement (replicability over time and with people)

Types of Scales/ Measurement Formats

Continuous Rating Scale

● Placing mark at appropriate part in the line

Itemized Rating Scales

● The respondents are provided with a scale that has a number or a quick description associated with each category

different categories:

Likert Scale ● Requires the respondent to indicate a degree of agreement or disagreement with each of a series of statements about an object/entity

Semantic Differential Scale

Designing Questionaires

Questionaire :a set of questions to measure constructs and record variables of interest

Reliability and Validity (Questionaire impacts data accuracy)

● It must communicate to the respondent what the researcher is asking ● It must communicate to the researcher what the respondent has said ● A sloppy questionnaire can lead to a great deal of in accuracy (reduces reliability and validity of the measurement)

Types of Questions ● Non structured ○ Open ended

● Structured ○ Close ended

Non Structured

No fixed responses

Structured Questions

Fixed response options

Structured questions can fall under two categories 1. Dichotmous (two response options to select from) a. Ex. yes or no questions 2. Multichromonous (multiple response options to select from) a. Ex. a list where u can choose many options

Response option sequence

● Natural or not natural sequence

Balanced vs Non Balanced Response Options

● Unbalanced response options would have biased responses

Response options content

Responses choices should be ● Collectively exhaustive ○ Taken together

● Mutually exclusive ○ Not overlap

● Not specific enough

Wording: Avoiding Jargons

● Avoiding unnecessary complicated language

Sequencing or Ordering of Questions in a Questionaire

Basics: Ordering Questions in a Quetionaire

Screening questions

● Are used to identify the respondents who qualify for the research study ● Screening questions are asked to ensure you are talking to your target popular ● Research objectives should specify who should and who should not be included in the research study

Arrangement of related questions

● Cluster questions that focus on the same topic ● Grouping questions into meaningful clusters can increase respondents ease in answering the questions and reduce the chance of respondents errors

Position of demographic and sensitive questions

● Respondents personal or demographic characteristics ○ Ex. age, education level, or income level are placed at the end of the questionnaire

● Questionnaire should look appealing and easy to work with

Funnel approach

● Approach to order questions in a questionnaire ● Narrow down questions from broad to specific

FINAL STEP - Pretesting ● Pretesting is administering a questionnaire to a limited number of potential respondents and other individuals capable of pointing out design flaws

Questionaire design important points

● Being with simple questions, then progress to harder and more complex ones ● Ask questions at the end ● Place sensitive questions at the end ● ** End with a thank you statement

Sampling

Process of collecting a sample is called sampling ● Selecting a small number of elements from a larger defined target group of elements ● Information gathered from the small group will allow judgements to be made about the larger groups

Why do we sample?

Simple Random Sampling

Taken from target population

Process 1. The sampling units in target population are numbered from 1 to 15 2. Random number generator randomly picks 4 numbers from 1 to 15 3. The numbers picked were ___ , ___, and ___ (3,2,10,14) 4. Sampling units with these numbers constitute the sample

Systematic Random Sampling

Stratified Random Sampling

● Sample selected based on researchers judgements/ beliefs

○ Ex: surveying sales people to understand problems customers face

because they interact w them the most

Snowball Sampling

● Each person interviewed is asked to recommend another person or additional

people for interviewing

○ Also called referral sampling

● Useful when sampling units are difficult to reach or rare

○ Ex. drug abuse, shoftlifting, etc

Quota Sampling

● Use convenience sampling to gather the sample and it is continued until different

strata of sampling units are represented in the sample in prespecified

proportions

● Looks similar to stratified sampling but instead of simple random sampling ,

convenience sampling is used for each stratum

Determind Sample Size (How many to sample?)

● Probability sampling

○ Sample size is found using a mathematical formula

○ The objective is to have a sample large enough to be confident that the

results will accuraltley reflect the population

● Non probability sampling

○ Sample size can not be determined using a mathematical formula

● Population parameters (mean and standard deviation) are generally unknown

● We estimate population parameter using sample statistics (mean and S)

EQUATION TO FIND SAMPLE SIZE (probability sampling)

● The difference between findings based on the sample and the true value for a

population

○ Happens when sample does not represent target population well

● Caused by sampling method used and size of the sample

● Can be reduced by

○ Increasing sample size

○ Using appropriate sampling method (ex. Probability sampling)

Non Sampling Errors

● All other errors other than sampling errors are non sampling errors

● Occur regardless of whether a sample or census is used

○ (census means target population)

● Can be two types

○ 1. Response Error

○ 2. Non response Error

Response Error

● When respondents have impared memory and do not respond accurately

○ Ex. do not remember last time they purchased a product

● Social Desireability Bias

○ When respondents respond in a socially desirable way (ex. I work out

every day)

Non Response Error

● A systematic bias that occurs when the sample differs from planned sample

● Occurs when a sufficient number of respondents in a preselected sample are not

able to participate

Multiple Choice

Multiple Choice

Was this document helpful?