MKT 343 exam 2

quantitative methods of collecting primary data …

quantitative methods of collecting primary data …

involve much larger samples. including survey designs used in descriptive and causal research. (EX

-How satisfied are our employees?→ descriptive

-Specifically looking at relationships among different things→ but if you look at if satisfaction leads to loyalty then that is causal

descriptive qualitative

research can be said to be descriptive on account of its vivid and detailed descriptions

descriptive quantitative

research is descriptive by using numbers and statistics to summarize demographics, attitudes, & behaviors

ex. mean, median mode, metrics etc. that this would be technically used as descriptive data

Advantages of Quantitative Survey Research Designs

• can accommodate large sample sizes so that results can be generalized to the target population

• produce precise enough estimates to identify even small differences

• easy to administer and record answers to structured questions

• facilitate advanced statistical analysis

• concepts and relationships not directly measurable ca be studied

Disadvantages of Quantitative Survey Research Designs

• questions that accurately measure respondent attitudes and behavior can be challenging to develop

• in-depth data difficult to obtain

• low response rates can be a problem

sampling errors:

• difference between the findings based on the sample and the true values for a population

• caused by a method of sampling used and the size of the sample

• can be reduced by→ increasing sample size and using the appropriate sampling method

respondent errors

• consist of both nonresponse error and response error

Nonresponse error

systematic bias that occurs when the final sample differs from the planned sample (b/c a sufficient number of sample respondents refuse to participate or cannot be reached)

Response error

When respondents have impaired memory or do not respond accurately

ex. for whatever reason the people who respond to your survey is based on some systematic reason, then you could be left with a response error.

Nonsampling errors characteristics

1. They create “systematic variation” (bias) in data

2. They are controllable

3. Cannot be directly measured (unlike random sampling error)

4. One nonsampling error can create others

end result → reduction of the quality of the data

●

Person Administered Survey Research

in-home interview or Mall-intercept interview

In home interview

an interview that takes place in the respondents home or in special situations, within the respondents work environment (in-office)

Mall intercept interview

Shopping patrons are stopped and asked for feedback during their visit to a shopping mall t

elephant administered survey research includes

traditional phone interview, computer assisted telephone interview, or wireless phone surveys

traditional phone interview

an interview takes place over the telephone. interviews may be conducted from a central telephone location or the interviewers home

computer assisted telephone interview (CAIT)

A computer is used to assist in a telephone interview

wireless phone surveys

wireless phones are used to collect data. The surveys may be text-based or web-based

Self Administered survey research includes

Mail survey, Online surveys, Mail Panel survey, and Drop-off survey

Mail survey

Questionnaires are distributed to and returned from respondents via the postal service or overnight delivery.

Online survey

The internet is used to ask questions and record responses from respondents

Mail Panel survey

Surveys are mailed to a representative sample of individuals who have agreed in advance to participate.

Drop-off survey

Questionnaires are left with the respondent to be completed at a later time. The surveys may be picked up by the researcher or returned via mail.

Advantages of Person-Administered surveys

Adaptability rapport, feedback, and quality of responses.

Adaptability report

trained interviewers can quickly adapt to respondents differences. Not all people are willing to talk to strangers when asked questions. Interviewers can establish a “comfort zone” during the questioning process and make the process of taking the survey more interesting to respondents.

Feedback

During the questioning process interviewers can answer respondents questions and increase the respondents understanding of the instructions and questions and capture additional verbal and non-verbal information

Quality of responses

interviewers can help ensure respondents are screened to represent the target population. Respondents are more truthful in their responses when answering questions face-to- face as long as questions are not likely to result in socially desirable biases.

disadvantages for person-administered surveys

possible recording error, interview respondent interaction error, and high expense

possible recording error

Interviewers may incorrectly record responses to questions.

interviewer- respondent interaction error

respondents may interpret the interviewers body language, facial, expression, or tone of voice as a clue to how to respond to a question.

high expense

Overall cost of data collection using an interviewer Is higher than other data collection methods.

Advantages of self administered surveys:

low cost per survey, respondent control, no interview respondent bias, anonymity in responses.

disadvantages of self administered surveys:

minimize flexibility, high nonresponse rates potential response errors, slow data acquisition, lack of monitoring capability.

situational characteristics include:

• –Budget

  –Completion time frame

  –Quality requirements

  –Completeness of data

  –Data generalizability: Projectable to the population represented by the sample in a study

  –Data precision

Data generalizability

Projectable to the population represented by the sample in a study

Task characteristics include:

–Task difficulty

–Required stimuli

–Amount of information asked from respondents

–Topic sensitivity

Topic sensitivity

The degree to which a survey question leads the respondent to give a socially acceptable response

What goes into selecting the appropriate survey method?

respondent characteristics, topic sensitive, and task characteristics

respondent characteristics

–Diversity

–Incidence rate

–Respondent participation

incidence rate

The percentage of the general population that is the subject of  the market research

respondent participation

ability, willingness, and knowledge level to participate

Ability to participate

The ability of both the interviewer and the respondent to get together in a question-and-answer interchange

Willingness to participate

The respondent’s inclination or disposition to share his or her thoughts

knowledge level

Degree to which the selected respondents feel they have knowledge of or experience with the survey’s topics

causal research

Studies that enable researchers to assess “cause-effect”  relationships between two or more variables- includes independent and dependent variables

independent variables

Variables whose values are directly manipulated by the researcher

dependent variables

Measures of effects or outcomes that occur as a result of  changes in levels of the independent or causing variable(s)

3 conditions for evidence of a “cause-effect” relationship (If X, then Y)

1.Temporal order between X and Y

2.Data confirm a meaningful association between X and Y

3.Must control for the impact on Y of any extraneous variables

correlation and causation relationship

Just because 2 things are related does not mean that one causes the other.

Experiment

An empirical investigation that tests for hypothesized relationships between dependent variables and  manipulated independent variables

Experimental designs collect data using both

experimental designs collect data using both survey and observation (quasi-experimental)

Nature of an experiment’s biggest challenge is

controlling for other influences

Control variables:

Do not vary freely or systematically with independent variables

–Should not change as the independent variable is manipulated

Extraneous variables:

Any variables that experimental researchers do not measure or control that may affect the dependent variable

Validity

The extent to which the conclusions drawn from an experiment are true

internal validity

Extent to which the research design accurately identifies causal relationships

external validity

Extent to which a causal relationship found in a study can be expected to be true for the entire target population

Laboratory (lab) experiments:

Causal research designs that are conducted in an  artificial setting

Field experiments:

Causal research designs that manipulate the independent variables in order to measure the dependent variable in a natural setting

–Performed in natural or “real” settings

Test Marketing

•Using controlled field experiments to gain information on specified market performance indicators

ex. Last part of the process -> see how people respond to the whole marketing mix of this product( experiment with products packaging and price

Sampling

–Selection of a small number of elements from a larger defined target group of elements

–Expecting that the information gathered from the small group will allow judgments to be made about the  larger group

Sampling as part of the research process

•Sampling is used when it is impossible or unreasonable to conduct a census

•The sample can influence research design/questionnaire

census

A research study that includes data about every member of the defined target population

Central Limit Theorem (CLT)

–The sampling distribution derived from a simple random sample will be approximately normally distributed (provided the sample is sufficient in size); sample mean will closely approximate population mean

–Key to understanding sampling error, statistical significance, sample size

•68% of observations lie within 1 standard deviation of mean

•95%of observations lie within 2 standard deviation of mean

Tools used to assess the quality of samples

sampling error and nonsampling error

Sampling error:

Any type of bias that is attributable to mistakes in either drawing a sample or determining the sample size

Nonsampling error:

A bias that occurs in a research study regardless of whether a sample or census is used

–Population frame error

–Inappropriate measures

–Questionnaire design

–Coding/data entry errors

2 difficulties with detecting the sampling errors:

–A census is very seldom conducted in survey research

–Sampling error can be determined only after the sample is drawn and data collection is completed

Probability Sampling

•Each sampling unit in the defined target population has a known probability of being selected for the sample

Nonprobability Sampling

••Sampling designs in which the probability of selection of each sampling unit is not known

••The representativeness of the sample cannot be measured

••The selection of sampling units is based on the judgment of the researcher and may or may not be representative of the target population

Probability sampling methods include:

simple random sampling, systematic random sampling, stratified random sampling, and cluster sampling

Nonprobability sampling methods include:

convenience sampling, judgement sampling, quota sampling, snowball sampling

Simple random sampling:

Every sampling unit has a known and equal chance of being selected (e.g., all names in a hat)

Systematic random sampling:

Similar tosimple random sampling but the defined target population is ordered in some way

–Usually in the form of a customer list, taxpayer roll, or membership roster, and selected systematically

–Allows the use of a skip interval to choose units

Steps for drawing a systematic random sample

1. obtain a list of potential sampling units that contains an acceptable frame of the target population elements

2. Determine the total # of sampling units making up the list of defined target populations elements and desired sample size

3. Compute the needed skip interval by dividing # of potential sampling units on the list by the desired sample size

4. using random number generator, randomly determine a starting point to sample the list of names

5. Apply the skip interval to determine remaining names that should be included in the sample

Stratified random sampling:

Separation of the target population into different groups, called strata, and the selection of samples from each stratum

•Based on income, location, etc.

–Proportionately stratified sampling

–Disproportionately stratified sampling

Proportionately stratified sampling:

  Each stratum is dependent on its size relative to the population

Disproportionately stratified sampling: 

The size of each stratum is independent of its relative size in the population

Cluster sampling:

Sampling units are divided into mutually exclusive and collectively exhaustive subpopulations, called clusters

–Area sampling

–Usually more specific than strata (e.g., customers on a given day, audience of a specific show)

Warning→ birds of a feather flock together

Area sampling

Clusters are formed by geographic designations

Judgment sampling:

Participants are selected according to an experienced individual’s belief that they will meet the requirements of the study

-Usually more common at earlier stages of research

Convenience sampling:

Samples are drawn at the convenience of the researcher

Quota sampling:

Participants are selected according to pre-specified quotas regarding demographics, attitudes, behaviors, or some other criteria

Snowball sampling:

A set of respondents is chosen, and they help the researcher identify additional people to be included in the study

–A.K.A. referral sampling

Factors to consider in selecting sampling design

research objectives, degree of accuracy, resources, time frame, knowledge of the target population, scope of the research, statistical analysis

•Factors that determine sample sizes with probability designs:

–Population variance and standard deviation

–More dispersion = higher sample size needed

–Level of confidence desired in the estimate

–Probability the sample is representative

–Degree of precision desired in estimating the population characteristic

•Precision

•Often, rules of thumb are used (along with budget/time                         

(e.g., n=250 for SEM)

Precision:

The acceptable amount of error in the sample estimate

Sample size formulas cannot be used for nonprobability samples

Determining the sample size is a subjective, intuitive judgment

Other Sample Size determination approaches

•Budget

•Similar previous studies with reliable, valid results

•Based on subgroups (50-100 units each)

Number of questionnaire items (5 units per item)

Measurement

•The process of developing methods to systematically characterize or quantify information about persons, events, ideas, or objects of interest

–Consists of two tasks:

•Construct selection/development

•Scale measurement

Construct

•An unobservable concept that is measured indirectly by a group of related variables

Construct Development

The process by which researchers identify characteristics that define the concept being studied

Consumer Concrete properties

age, sex, marital status, income, brand last purchased, dollar amount of purchase, types of products purchased, color of eyes, hair.

Consumer abstract properties

attitudes toward a product, brand loyalty, high-involvement purchases, emotions (Love, fear, anxiety), intelligence, and personality

organizations concrete properties

name of company, number of employees, number of locations, total assets, Fortune 500 rating, computer capacity, types and numbers of products and service offerings.

organization abstract properties

competence of employees, quality control, channel power, competitive advantage, company image, consumer-oriented practices.

Scale measurement

•The process of assigning descriptors to represent the range of possible responses to a  question about a particular object or construct

Scale points

Designated degrees of intensity assigned to the responses in a given questioning or observation method

4 scale levels

nominal, ordinal, interval, ratio

Nominal Scale (most basic, least powerful: report counts, mode)

•The type of scale in which the questions require respondents to provide only some type of descriptor as the raw response

Ordinal Scale(report mode, median, mean, frequencies, and range)

•A scale that allows a respondent to express relative magnitude between the answers to a question