ADV 309R Exam 1 Study Guide - Dr. Oh

Advertising & PR Research

Definition of Advertising

Persuasive Communication (promotes products, services, or causes)

Mass-Mediated (non-personal, monologue)

Paid for (time or space in the media)

Primary Research

Data you collect yourself, proprietary information

Secondary Research

Data collected from existing sources, often used to complement primary research.

Qualitative Data

non-numerical information gathered through methods like interviews and focus groups to understand consumer attitudes, motivations, and behaviors

Quantitative Data

Data revolving around numbers to provide generality when working with a large population.

How does advertising research contribute to the understanding of market, consumer, creative, and media?

Market: market trends, competitive environments
Consumer: attitudes toward brand, lifestyle, needs/wants, Consumer segments (millennials/Hispanics)
Creative: Brand Positioning, Concept Testing, emotional response
Media: What platforms would be best to advertise on, what platforms does the target audience spend time on?

Steps in Research Panning and Data Collection

1. specify TYPE of research (primary or secondary)
2. Determine Sampling Plan
3. Select Data Collection Method

Principles of Research Ethics

1. Nonmaleficence- not doing harm or bad on respondent. No lying

2. Beneficence - Doing a good or nice thing for respondent

3. Autonomy - Respondent has informed consent and is in control of their participation in the survey, decision power, not the researcher

Responsibilities to respondents

1. Informed Consent - context of research, voluntary, how they will keep data confidential
2. Mistreatment - No physical or psychological mistreatment
3. Confidentiality and Privacy - respondents data is confidential, no identifiable information leaked

“Three Identical Strangers” Ethical Issues

• informed consent

• privacy issues

• psychological harm

• violates all three forms of privacy

Target Population

The specific group of individuals that a study aims to understand or draw conclusions about, which the research findings will be relevant to.

Census

A complete count of a population, typically conducted at regular intervals to gather demographic information.

Probability Sampling

A sampling technique ensuring each member of the population has a known, non-zero chance of selection, allowing for greater result generalization.

3 Types of Probability Sampling

Simple Random, Systematic, Stratified Sampling

Simple Random Sampling

each member of the population has an equal chance of being selected, often achieved through random number generation.

Systematic Random Sampling

Randomly generate a starting point in a population then calculate the skip interval by dividing the population size by the desired sample size, then with the calculated number, select every nth member.

Stratified Random Sampling

Used in non-homogenous populations to ensure that different sub-groups are represented. This is done by putting subjects into strata, then carrying out a simple random sample within each stratum.

Proportionate Stratified Sampling

A method that guarantees representation of various sub-groups in the sample, maintaining the initial ratio of each stratum in relation to the population.

Disproportionate Stratified Sampling

intentionally selects different subgroups at varied rates to provide deeper insight into smaller, underrepresented populations. Can change initial stratum ratios from 2:3 to 3:3 to gather more coverage of the smaller group, thus throwing off the original proportion

Nonprobability sampling

not all population members are selected, leading to potential bias and limited generalizability.

3 Types of Non-Probability Sampling

convenience, judgment, quota, and snowball sampling - based on subjective selection.

Convenience Sampling

method where subjects are selected based on their easy availability and proximity to the researcher.

Judgement Sampling

subjects are chosen based on the researcher's discretion, relying on their judgment rather than random selection.

Quota Sampling

researchers ensure specific subgroups are represented in the sample based on predetermined quotas.

Snowball Sampling

existing study subjects recruit additional subjects from their networks.

Survey Methods

• email

• telephone

• personal interview

• online

Response Rate

The percentage of respondents who complete a survey out of the total number invited or selected.

Calculated by dividing the number of respondents completing the survey by the number of respondents in the valid sample.

Nonresponse Rate

measures the proportion of individuals in a survey who did not respond compared to the total number of individuals selected for the study.

calculated by dividing the number of nonrespondents by the total number of potential respondents, expressed as a percentage.

Nonresponse Error

Refusals - people declining or failing to complete survey
Not-at-homes - people unavailable during contact window

The Harvard Study Example

Longitudinal Study (followed the men throughout their whole lives)

Results found that healthy and well-maintained relationships are the key to a happy and long life

Measurement

Linking abstract, unobservable concepts to observable events

Concept

A name for a property of an object, person, state, or event.

Ex. Bravery (property of a person), Love (feeling, virtue, advertising, b rand attitude

Index

When several questions are used to measure one concept (often uses likert scale for 5-6 categories)

Operational Process

The process of developing measurement tools for concepts

• Uncover possible meanings by brainstorming

• State/find major characteristics

• Theoretical (conceptual) definition

• Operational definition

Theoretical Definition

summarize the central or core idea of the concept in a sentence

Operational Definition

translate the theoretical/conceptual definition into more observable specific evaluations

Four Levels of Measurement

Nominal, Ordinal, Interval, Ratio

Nominal

Yes or No, Likely or Not Likely, Like or Dislike questions to measure something, numbers assigned to different options have no meaning

Ordinal

Rank 1-5, order matters, no assumed equal distance between numbers

Interval

Likert’s Scale, no true 0, Cannot measure a ratio, assumed equal distance between points

Ratio

I trust him ____% (0% - 100%)

True 0 point

Can measure equal distance between each point

can make an accurate ratio from data

Ways of Assessing Reliability

Test-Retest - repeats test item under equivalent conditions to the same group of people

Internal Consistency - answers of the question should all converge toward the same direction across many respondents

Ways of Assessing Validity

Face Validity
Concurrent Validity
Predictive Validity

Face Validity

whether a test looks like it measures what it's supposed to measure, based on a quick assessment.

arousal scale

intuitively makes sense

Concurrent Validity

measures how well a new test matches up with an established test taken at the same time. If both tests give similar results, the new test has good concurrent validity.

Likert scale using operationalization to further solidify subjects conceptual reaction

Predictive Validity

• how well a measure, like a customer survey, can forecast actual holiday sales. If the survey predicts high sales and that happens, it has good predictive validity.

• arouse consumer (negative/positive) more likely to to take action

  • Happy and excited consumers during xmas are more likely to make a purchase from an xmas sale

Reliability of Measurement

measures the extent to which a measure consistently provides same results

Validity of Measurement

assesses how accurately a measure actually measures what we think it is measuring