Exam 2

Introduction a. What type of information should it include?

What is the survey about? indicate to respondents how they were selected. Incentives are offered for participation Ensuring anonymity and confidentiality reduces anxiety

What are warm up questions?

are simple and easy-to-answer questions that are used to get the respondents' interest and to demonstrate the ease of responding to the research request.

Scale measurement (unit)

he process of assigning a set of descriptors to represent the range of possible responses to a question.

What are screening questions?

used to identify respondents who do and who do not meet qualifications necessary to take part in the research study.

What are classification questions?

(typically demographics) are used to classify respondents.

Test Markets

- Consists of controlled field experiments - Conducted in limited market areas - Specific market performance indicators - Customer attitudes towards a new product, idea, service, or communication strategy

Correlation v. Causation

- Correlation = relationship between two variables - Causation = one variable producing an effect on another variable - Correlation ≠ Causation

What are surveys? - Advantages

- Must have a good question - standardization - gathers large amounts of information - easy to administer - easy to analyze - reveal subgroup differences

What is questionnaire organization? a. What are the 5 typical sections types to include? b. Know how to organize questions using these 5 sections.

- Question Organization pertains to the sequence of statements and questions that make up a questionnaire. 1. Introduction 2. Screening 3.Warm-Up 4.Transition 5. Classification

What is a questionnaire?

- is the vehicle used to present the questions the researcher desires respondents to answer. - The researcher should strive to minimize question bias. -Question bias is the ability of a question's wording or format to influence respondents' answers.

What are the four steps of questionnaire development?

1. Determine what and how to measure Identify construct's properties and type of measure for each 2. Decide on wording Word each question using dos and avoiding donts 3. Organize questionnaire Write introduction, put questions in order, skip logic as needed 4. Finalize and launch get client approval, pretest, code, launch

What is panel company? Advantages/disadvantages

A panel company recruits a large number of potential respondents who agree to take part of a survey for compensation. ADVANTAGES Fast turnaround High quality Database information Access to targeted respondents Integrated features DISADVANTAGES Not random samples Overused respondents Cost

Census Study vs. Sample Study

Census - nferences from the entire body of units of interest Sample - inferences from a sample drawn from the population

Survey classification factors: - Is an interviewer used? (what are the two types) - Is a computer used? (what are the two types)

Computer v. Interviewer - Interviewer person administered or self administered - Computer computer assisted or computer administered

What is the difference between cross sectional and longitudinal studies?

Cross - describes the population at one point, sample surveys Long - same sample is measured repeatedly over time, panel studies

How do we manage extraneous variables?

Design Controls - only differences between groups is manipulated e.g. same location, same time, same day. Randomization - random assignment of test units to groups - ensures prior equality of experimental groups

Types of problems, questions, methods compared to other research types

Designs - Exploratory > qualitative [emphasis on the discovery of new ideas] - Descriptive > quantitative Methods [Concerned with determining the frequency with which something occurs or the relationship between two variables] - Causal > quantitative [Concerned with finding cause-and-effect relationships] - cross sectional v. longitudinal

Elements, frames, examples of each

Elements - who we will collect data from. Population - All elements added together Sampling Frame - List of all possible sampling units

What are lab and field studies? a. Pros and cons of field v. lab studies

Field - Natural setting but extraneous variables cannot be controlled - High internal validity, high scientific control Laboratory - Artificial setting where extraneous variables can be controlled - high external validity, real world environment

What are three things to consider when selecting a data collection mode?

How fast is data collection? How much will data collection cost? How good is the data quality?

Person Administered surveys - Advantages and Disadvantages - In home interview, mall-intercept interview, in-office, telephone

In home - conducts the interview in the respondent's home, normally at a preset appointment time. Mall - Shoppers in a mall are approached and asked to take part in the survey. Questions may be asked in the mall or in the mall-intercept company's facilities located in the mall. Office - The interviewer makes an appointment with business executives or managers to conduct the interview at the respondent's place of work. Telephone - Interviewers work in a data collection company's office in cubicles, usually reading questions on a computer monitor. Often the supervisor has the ability to "listen in" to interviews in order to check that they are being conducted correctly. ADVANTAGES - Feedback Rapport Quality control Adaptability DISADVANTAGES - Humans make errors Slow speed High cost Fear of interview evaluation

Identify and know definition for independent, dependent and extraneous variables.

Independent - Values manipulated by researcher Dependent - Measured of effects/outcomes Extraneous - Other variables that may affect the dependent variabledependent variable

What is internal vs. external validity in experiments?

Internal - the extent to which the research design accurately - identifies causal relationships. External - he extent to which a causal relationship is true for the defined target population.

A/B testing

Is a way of comparing two versions of something to figure out which one performs better. It is an experiment!

now/Identify/Write some commonly used Marketing scales a. Likert b. Semantic differential c. Behavior intention scale d. Graphics e. Constant Sum scale **examples**

Likert - scale asking to what extent you agree or disagree with a series of statements. - Strongly disagree to strongly agree Semantic - bipolar scale that captures a person's attitude - For me vs. Not for me Behavior Intention - A scale which captures the likelihood someone will perform a behavior - I definitely will buy Graphics - The hospital pain scale with the faces Constant Sum - A scale where respondents allocate points among different options. - Please allocate 100 points among the brands so that the allocation represents your preference of each brand. Points should add up to 100.

Estimating sample sizes (be able to do calculations). I will provide the formula. a. For means b. For proportions c. Tolerance, confidence, standard deviation

Mean - Calculation without a "p" value Proportion - calculation with a p value Tolerance - T in the formula, given Confidence - z score - 99 = 2.575 - 95 = 1.96 Standard Deviation - or range/6

What is causal research? What does it aim to do? What is the methodology?

Method: - Experiments Characteristics: - Manipulation and control of variables Purpose: - Determining cause and effect Goals and Objectives: - cause-effect" relationships - If X, then Y - Controls

Levels of measurement - Know the definition, properties of each one. Be able to write, identify questions at different levels. a. Nominal b. Ordinal c. Interval d. Ratio ** examples included **

Nominal (assignment) - labels only, no ordering - individual is one group or another - Numbers assigned to ice skaters Ordinal (order) - Indicate relative size difference between items - X is better than Y - Rank order of winners Interval (distance) - the exact difference between scale point descriptors. (Interval level scale) - X is _____ units different than Y - performance evaluation on a scale of 0 to 10 Ratio (origin) - has a true zero point. (Ratio level scale) - Specific # - Time to finish program

What is a nonprobability sample? What is a probability sample? a. Nonprobability designs b. Convenience c. Judgment/ snowball sample d. Quota

Non Probability - Rely on personal judgments to determine who will participate in the study. - There is no way to ensure that the sample is representative of the population. - Should be used with caution. Convenience - Intercepts - Problem- Generalizability? Judgement - Sample respondents are selected because the researcher thinks they meet the requirements of the study Quota - samples attempt to ensure that the sample is representative by selecting sample elements in such a way that the proportion possessing a certain characteristic is approximately the same as the proportion of the population.

Self-administered surveys - Advantages and disadvantages - Online survey, Group self-administered survey, drop-off survey, mail survey

Online - Respondents answer a questionnaire that resides on the Internet. Group-Respondents take the survey in a group context. Each respondent works individually, but they meet as a group, which allows the researcher to economize. Drop Off - Questionnaires are left with the respondent to fill out. The administrator may return at a later time to pick up the completed questionnaire, or it may be mailed in. Variations include surveys handed to customers or placed where respondents will naturally encounter them, such as in a hotel room. Mail - Questionnaires are mailed to prospective respondents, who are asked to fill them out and return them by mail. ADVANTAGES - Reduced cost Respondents can control the pace. Reduced interview evaluation apprehension DISADVANTAGES- Cannot control if respondents will complete the survey, skip questions, etc. Lack of monitoring The questionnaire has to be perfect.

Populations v. Samples

Population - Population: Entire group of interest that we are trying to study. - group taken from population

Question ordering a. Purpose of question ordering b. Effects of poor ordering = bias

Purpose is to build and flow from one question to another

What's reliability? What's validity?

Reliability - respondent responds in the same or a similar manner to an identical or nearly identical measure Validity - accuracy or exactness of the measurement

Probability designs a. Simple random samples b. Systematic c. Cluster d. Stratifide i. Proportionate ii. Disproportionate

Simple Random Sampling - Each element of the population is given an equal probability of being selected. - Uses random numbers to select the elements. Systematic Sampling - The elements are ordered according to some criteria and elements are selected at regular intervals through the ordered list. - Involves a random start and then proceeds to select every kth element from that point onwards. - Skip Interval Cluster - Usually refers to geographic areas - The researcher first randomly selects a couple of clusters and then measures all the units within that cluster. Stratified - The elements are divided into homogeneous and non-overlapping groups (called "Strata") and the simple random sample is drawn from within each group. 1. Proportionate (representative of highschool classes) 2. Disporportionate (25% of each)

Dos/Don'ts of measurement a. Identify what is wrong with a question. b. Be able to rewrite it correctly c. Problems with Ambiguous Wording

The 4 Do's - The question should be focused on a single issue or topic. - The question should be brief. - The question should be grammatically simple. - The question should be crystal clear. Leading - A leading question gives a strong cue or expectation as what answer to provide. - Shouldn't concerned parents of toddlers use infant car seats? Loaded - A loaded question has buried in its wording elements a sneaky presupposition. - Do you always cheat in exams? Double Barreled - A double-barreled question is really two different questions posed in one question. - Do you like candy and Paris? Overstating - An overstated question places undue emphasis on some aspect of the topic. - How much do you think you would pay for a pair of sunglasses that will protect your eyes from the sun's harmful ultraviolet rays, which are known to cause blindness?

What is your goal when you are collecting data with a questionnaire?

To answer the research objectives and problem statement - collect needed data - minimize error - foster cooperation/completion

What is a scale?

a device that includes a range of values

What is a pretest?

a dry run of a questionnaire to find and repair difficulties that respondents encounter while taking the survey.

Factors necessary for causality a. Artificial correlations (spurious) b. Be able to explain why something is causation or correlation. c. Be able to identify "third variables" or alternative explanations

a. statistically real but not meaningful Danger of data mining b. People buy more ice cream when it's hot (causation) sharks attacks and ice cream increase in the summer (coorelated) c. Third party factors (ice cream does not cause shark attacks, warm weather does)

How do changes in each of these impact required sample size?

causes increases or decreases in # of samples needed to meet the experiment needs

Scale points

the designated degrees of intensity assigned to the responses of a question.

What is measurement?

the process of describing or assigning intensity to information about things