ADV 281 Exam 1 Study Notes

Pierce's Four Paths

  • Worst to Best:

    1. Method of Tenacity

    2. Method of Authority

    3. A Priori Method

    4. Scientific Method

Method of Tenacity

  • Defined as clinging to familiar beliefs because they provide peace of mind.

Method of Authority

  • Based on the premise that someone who is older, wiser, or more experienced (possibly holding a title) informs what is right or wrong.

  • Example: The Pope declaring the Earth as the center of the universe, which led to widespread belief.

A Priori Method

  • Involves reasoning from cause to effect.

  • Entails careful thought and self-reflection in deciding matters.

Scientific Method

  • Centers on reliance on empirical inquiry, which must be observable by anyone.

  • Characterizes by specific rhetoric involving independent and dependent variables.

  • Recognizes limits and boundaries to scientific truths, asserting that they are probabilistic.

  • Involves systematic techniques or procedures for analyzing empirical evidence in an unbiased attempt to confirm or disprove prior conceptions.

Scientific Method (Step by Step)

  1. Observation/Question

  2. Research Topic Area

  3. Hypothesis

  4. Test with Experiment

  5. Analyze Data

  6. Report Conclusions

Basic vs. Applied Research

  • Basic Research: Conducted to enhance knowledge about a topic.

    • Example: Investigating visual attention in 4-6 year old children, often conducted in a lab.

  • Applied Research: Aimed at solving practical problems with findings that can be directly applied to real-world issues.

    • Example: Investigating effective methods for teaching reading to children, often conducted in the field.

Lab vs. Field Research

  • Lab Research:

    • Has fewer confounds; greater control in study design.

    • May be artificial in nature.

  • Field Research:

    • May contain more confounds.

    • Researchers have less control over variables.

Inductive vs. Deductive Reasoning

  • Inductive Reasoning:

    • Involves developing generalizations from specific observations.

  • Deductive Reasoning:

    • Entails developing specific predictions from general principles.

Quantitative vs. Qualitative Research

  • Quantitative Research:

    • Measures differences in the amount of behavior exhibited.

  • Qualitative Research:

    • Describes differences in the kind or quality of behavior shown.

Basic / Lab Research Characteristics

  • Knowledge-driven

  • Lab-based

  • Experimental approach

  • Nomothetic (aimed at broad generalization)

Field / Applied Research Characteristics

  • Problem-driven

  • Field-based

  • Correlational approach

  • Idiographic (very narrowly focused)

Possible Ways to Measure Reading

  • Methods:

    • Self-report

    • Observation

    • Existing data (archives or secondary data)

Self-Report

  • Not ideal due to limitations in human memory.

  • Particularly problematic for sensitive subjects.

  • Example: Asking individuals how much they read newspapers may yield inaccurate responses.

Observation

  • Challenging because it does not confirm whether people are actively engaged or just glancing at the material.

  • Requires willingness from people being observed or must occur in public spaces.

  • Example: Observing behavior in locations where newspapers are present.

Existing Data "Archives" or Secondary Data

  • Example: Analyzing leftover newspapers.

Advertising-Related Concepts

  1. Advertising Efficacy:

    • Questions regarding how we assess the effectiveness of an advertisement.

  2. Brand Equity:

    • Refers to the strength of a brand in relation to competing brands.

  3. Product Use:

    • Noting that a product might be purchased but never utilized.

  4. Brand Loyalty:

    • Can be defined or conceptualized in numerous ways.

Conceptual Definition

  • A verbal explanation delineating the meaning of a variable.

  • Clarifies what the concept encompasses and what it excludes.

  • Resembles a dictionary definition.

  • Example: Defining "attitude" which is a tendency that may change over time.

  • Indicators: Metrics to assess various dimensions of an attitude.

Aaker's Brand Personality Dimensions

  1. Sincerity

  2. Excitement

  3. Competence

  4. Sophistication

  5. Ruggedness

Operational Definition

  • A detailed description of how something is measured.

  • Specifies the operations necessary to obtain data.

  • Can involve observable indicators such as a yawn that signals tiredness.

Measures of Central Tendency

  • Mean: Average

  • Median: Middle value

  • Mode: Most frequently occurring value

Measures of Dispersion

  1. Range: Distance between the smallest and largest value in a dataset.

    • Formula: Range = Largest Value - Smallest Value

  2. Variance: Measurement of how far each number in the dataset is from the mean.

    • Expressed in squared units.

    • Relationship to Standard Deviation: Standard deviation is the square root of variance.

Conceptualization vs. Operationalization

  • Conceptualization: The process of defining a variable.

  • Operationalization: The methodology for measuring that concept.

  • Example:

    • Age can be conceptualized as the total years lived since birth.

Normal Distribution

  • Described as a bell-shaped curve.

  • Almost all values within the distribution fall within plus or minus 3 standard deviations.

  • Example: IQ scores typically reflect normal distribution.

Ethics

  • Defined as moral principles that guide the determination of right or wrong actions.

Deontology

  • A branch of knowledge focused on moral obligations.

  • Asserts that certain actions are inherently immoral or unethical regardless of consequences.

Two Critical Principles of Deontology

  1. Rights Principle:

    • Evaluated based on universality and reversibility (treat others as you wish to be treated).

  2. Justice Principle:

    • Distributive Justice: Fair distribution of benefits based on contribution.

    • Retributive Justice: Makes amends for harm caused.

    • Compensatory Justice: Ensures fair compensation for injuries.

Utilitarianism

  • Focuses on the consequences of actions, aiming for "the greatest good for the greatest number."

Common Concerns in Codes of Ethics

  1. Respect for Persons: Ensuring inform consent and protecting individuals with diminished autonomy.

  2. Beneficence: Striving to maximize benefits while minimizing harm during research.

  3. Justice: Identifying who deserves the benefits of research and who should bear its burdens.

Ethical Issues in Researcher-Participant Relationship

  1. Voluntary participation.

  2. No harm to participants.

  3. Anonymity and confidentiality.

  4. Informed consent.

  5. Minimizing deception in the research process.

Confidentiality and Privacy

  • Confidentiality is absolute unless participants are informed otherwise.

Research Question vs. Hypothesis

  • Research Question: Widely framed inquiries that are broad in scope.

  • Hypothesis: Specific predictions articulated as statements to be tested.

Levels of Measurement

  1. Nominal Scale:

    • Uses numbers or letters as labels for identification or classification without meaningful numeric representation.

    • Appropriate only for mode as a measure of central tendency.

    • Example: User/nonuser or occupations.

  2. Ordinal Scale:

    • Arranges objects according to their magnitude in an ordered relationship, indicating relative size without precise distances.

    • Example: Football rankings.

  3. Interval Scale:

    • Organizes objects by magnitude and distinguishes them using equal unit distances.

    • Example: Temperature or GPA, applicable for mean, median, and mode.

  4. Ratio Scale:

    • Reflects absolute quantities, possessing an absolute zero point representing absence of an attribute.

    • Example: Units sold or income, applicable for mean, median, and mode.

  5. Higher Levels of Measurement:

    • Higher scales encompass the properties of lower levels (e.g., Ratio includes Nominal, Ordinal, and Interval properties).

Criteria for Good Measurement

  1. Reliability

  2. Validity

  3. Sensitivity

Reliability & Assessment

  • Defines as consistency in results over time.

  • Assessing Reliability:

    • Test-Retest Reliability: administering the same test to the same group under equivalent conditions.

    • Alternative Form: offering different forms of the same test to the same group at two separate times.

Validity & Assessment

  • Indicates that a measure assesses what it is supposed to measure.

  • Reliability must precede validity.

  • Assessing Validity:

    • Face Validity: Consensus agreement among experts.

    • Concurrent Validity: Comparison with accepted measures.

    • Predictive Validity: Ability of the measure to forecast future outcomes.

Sensitive Measure

  • Captures subtle changes in the target response accurately.

Cross-Sectional Surveys

  • Provide insights on whether two variables are correlated.

  • Limitations:

    • Cannot ascertain a cause-and-effect relationship.

Establishing Cause-and-Effect Relationships

  1. Covariation: There is a relationship between cause and effect.

  2. Temporal Precedence: The cause precedes the effect chronologically.

  3. Internal Validity: Ability to rule out plausible alternative explanations.

Conceptual & Operational Definitions

  • Must align consistently with one another to ensure clarity in research.

Sample

  • The group of all respondents in a study.

Subjects/Respondents/Participants

  • Terms that can be used interchangeably to describe individuals involved in a study.

Sampling

  • The process involved in selecting the respondents for research.

Representative Sample

  • A sample reflecting the larger population, selected using statistical random techniques.

Population

  • The larger group that researchers intend to study.

Non-Representative Sample

  • Does not accurately represent all respondents across the target population.

Applying Conceptual/Operational Definitions to Cassandra's Example

  • Conceptual: Defines "reading the newspaper" in terms of minutes spent engaged with it per day.

  • Operational: Must correlate with how it will be measured effectively.

Reasons Samples Fail to Represent a Population

  1. Bias: Samples may inadvertently favor a particular type of participant.

  2. Error: Random chance discrepancies in representation.

Archives

  • A collection of existing resources (information or data) available for research purposes.

  • Note: Notably includes other forms of data beyond just numerical information.

Empirical Questions

  • Questions answerable through direct observation of the world; can yield qualitative or quantitative data.

Histogram

  • A graphical representation of score frequencies for a variable.

  • Plotted with the variable's scores along the x-axis and frequency counts on the y-axis.

Z-score

  • Indicates how far an individual score lies from the mean in terms of standard deviations, with positive indicating above the mean and negative indicating below the mean.

Directional vs. Nondirectional Hypothesis

  • Directional Hypothesis: Specifies the direction of the expected relationship.

  • Nondirectional Hypothesis: Does not specify the direction of the association between variables.

Common Mistakes in Hypothesis Testing

  • Testing hypotheses by merely asking if people believe the hypothesis to be true, rather than through experimental or observational methods.

Measurement Levels Classification

  • Categorical: Number of categories not arranged in any order.

  • Ordinal: Categories are ordered but not indicative of exact distances.

  • Interval: Numerous categories indicating specific amounts arranged from one extreme to the other.

Levels of Measurement Examples

  1. Biological Sex: Categorical/Nominal

  2. Time Taken to Run 100 Meters: Interval

  3. Computer Operating System Type: Categorical/Nominal

  4. Feelings Today: Ordinal (unhappy, meh, happy)

  5. Support for Increased Defense Spending: Ordinal (very low to very high)

Functions of Theory

  1. Provides insights on necessary observations to validate the theory.

  2. Acts as a repository for research aiding in building or modifying existing theories.

Correlation

  • A statistic measuring the strength of association between two variables, ranging from -1 to 1.

Effect Size

  • Describes the strength of the relationship between two variables.

Confidence Level

  • Denotes the probability that the true population correlation coefficient lies within a confidence interval.

  • Note: Wider confidence intervals usually signify higher confidence.

Null Hypothesis

  • Represents the assumption that there is no significant difference between specified populations, attributing any observed differences to sampling or experimental errors.

Statistical Significance

  • Indicates that it is highly improbable that the true population value of a correlation is zero, suggesting that the null hypothesis is false.

Confidence Interval

  • A range of values surrounding the observed sample mean.

  • As sample sizes increase, confidence intervals tend to narrow.

  • If a confidence interval does not include zero, the correlation is statistically significant.