COM 309 Final Exam

Four essential characteristics present for hypothesis

1. Compatible with current knowledge in the area 2. Logically consistent 3. Stated concisely 4. Should be testable

What is an analytical survey?

-Attempts to describe and explain why situations exist. -Two or more variables are usually examined to investigate research questions or test research hypotheses. -Examine interrelationships among variables and develop explanations

What is a descriptive survey?

-Attempts to describe or document current conditions or attitudes -Explain what exists at that moment -Dept. of Labor conducting study to determine rate of unemployment → EXAMPLE

Key elements of an experiment

-Manipulation of independent variable -Create conditions that isolate cause -Measure the effect (DV) after exposure to the cause (IV) -Control group gets no treatment -Experiment group gets treatment -Random assignment of participants

What is a sampling frame?

-The complete list of members in the population -Lists that include all members of a population have a high degree of precision

8 threats to validity

1) History: external event occurs that affects the results of the study 2) Maturation: respondents change over time, which affects the DV 3) Experimenter bias: existing attitudes of researcher affects results 4) Sensitization/Testing: participant is familiar with the measure (increases scores) 5) Regression to the mean: people may score lower/higher first, but then score average second 6) Contamination: people who have been part of study tell others about it 7) Sample bias/non-equivalent groups: conditions not equal before starting study

How many people won't show up for a focus group

1-2

Two Major Research Sectors

1. Academic (basic) college students conduct public sector research and have a theoretical or scholarly approach. Attempt to explain mass media effects on individuals 2. Private (applied) nongovernmental companies intended to facilitate decision-making, media content, and customers preferences

Three disadvantages of lab experiments

1. Artificiality 2. Researcher (experimenter) bias 3. Limited Scope

Four ways of knowing

1. Authority- Comes from a trusted source, something that "should know" 2. Tenacity/Tradition- You've always done it 3. Intuition/Logic- Self-evident 4. Scientific Method

Four keys to ethics

1. Autonomy of participant- Can drop out if they want 2. Nonmaleficence- Doesn't harm others 3. Beneficence- Remove potential harms 4. Justice- Treat everyone with respect

Question Guidelines

1. Be clear (avoid jargon or slang, write for a 9th grade audience) 2. Keep questions shorts 3. Avoid negatives 4. Avoid double-barreled questions 5. Avoid leading/biased questions 6. Avoid asking difficult to answer questions 7. Avoid false premises (don't assume) 8. Avoid asking embarrassing questions unless necessary 9. Response options should: match the question, be mutually exclusive, be exhaustive, have equal number of positives and negatives

Five interview issues

1. Be clear about motives 2. Anonymity 3. Final say-let participants look at the work 4. Money- are you going to pay people 5.Logistics- set up schedule

What is a way to cut breakoffs by 50%?

1. Cash (no less than $10) 2. Gift cards to local retailers 3. Prepaid telephone calling cards activated when questionnaire is returned 4. Drawing that offers a large, tangible prize

Criteria for causation

1. Cause & effect must be correlated 2. Cause must precede the effect 3. Study must account for alternative explanations

Important steps in field observation

1. Choose the research site 2. Gain access- contact and get "in" with groups 3. Sampling- decide # of groups/settings to observe 4. Exiting- have a plan to exit if you're doing covert research

Four conditions for interviews

1. Clear interests 2. Look at past events or difficult to study situations 3. Short time frame 4. Lots of perspectives

Four Limitations of Content Analysis

1. Content analysis can't serve as sole basis for claims about effects of content on an audience 2. Findings of content analysis are limited to the framework of the categories/definitions used in the analysis 3. Lack messages relevant to the research 4. Content analysis is frequently time consuming and expensive

7 steps for developing a focus group

1. Defining the problem- what are you talking about 2. Sample selection- who will participant 3. Determine # of groups- 6-10 people per group. Want multiple groups with different characteristics 4. Study Mechanics- where/when will it be held 5. Prepare materials- protocol, consent forms 6. Conduct the group 7. Analyze data

Three types of field observation notes

1. Descriptive- describe situation 2. Analytic- your analysis using theory; expand upon later that day 3. Autobiographical- your own behaviors & emotions

10 steps associated with conducting a content analysis

1. Develop a research question 2. Define the problem-narrow down question 3. Select a sample- multi-staged cluster most common. Want sample to representative 4. Define unit of analysis 5. Develop content categories- concepts must be measurable. Should be exhaustive, reflect purpose of research, and mutually exclusive 6. Establish quantification system (nominal interval) 7. Train coders-reliability is goal 8. Code data- apply coding scheme and generate data 9. Analyze the data 10. Draw conclusions

Issues with Variance

1. Doesn't make sense to say average error in data squared 2. Not tied to the measurement you're using

Four advantages of laboratory experiments

1. Evidence of causality 2. Control 3. Cost 4. Replication

Two things that very in field observation

1. Extent to which you're involved with the group 2. Extent to which the group knows you're conducting research

Five Assessments of Validity

1. Face Validity (looks good at face value) 2. Content Validity (measure captures the full ranges of meanings and dimensions of the concept) 3. Criterion/Predictive Validity (measure is shown to predict scores on some relevant future measure 4. Concurrent Validity: Measure discriminates between groups that should differ on key variable 5. Construct Validity: The measure is shown to be related to other concepts that should be related

Types of Descriptive Statistics

1. Frequency- # of times a given response is reported 2. Central Tendency- Point in distribution where data is centered (similarity between answers) 3. Dispersion- How spread out the data is, how different peoples responses are

Broad Leadership Traits

1. Group goal facilitation- helps group attain its goals 2. Group sociability- keep the group functioning smoothly 3. Individual prominence- desire for group recognition

Two questions researchers must learn to answer

1. How to use research methods & statistical procedures 2. When to use them

What are the four disadvantages of survey methods?

1. Independent variables cannot be manipulated the same as in lab experiments 2. Inappropriate wording or placement of questions can bias results 3. Wrong respondents may be included in survey research. → people can lie 4. Response rates continue to decline

Survey Introduction

1. Introduce yourself 2. Introduce your purpose 3. Risk Benefits, time required 4. Emphasize you're not selling anything

Issues with picking a research topic

1. Is topic is too broad 2. Can problem be investigated 3. Can data be analyzed 4. Is the problem significant 5. Can results be generalized 6. What cost/time is involved 7. Is there any potential harm

7 types of other focus group questions

1. Leading-push in a direction 2. Testing- test limits of what group will discuss 3. Steering- talk about uncomfortable topics 4. Feel- expose personal feelings 5. Anonymous- fill out notecard 6. Factual- less risk for person to answer 7. Silence

3 types of in-depth interviews

1. Life History- autobiographies that can help us understand social theories 2. Event-centered- Understanding what the researcher can't directly observe 3. Situational- Understanding a situation from multiple settings

Four research phases in mass media

1. Medium itself (what is it/how does it work) 2. Uses/Users of the medium 3. Effects of the medium 4. How it can be improved

Measurement Components

1. Numerical-no implicit quantitative meaning 2. Assignment- designation of numerals/numbers to certain objects or events 3. Rules- specify the way that numerals/numbers are to be assigned

Systematic Random Sample (Probability Sampling)

1. Obtain a list of all population members 2. Assign numbers to all members 3. Randomly select start position in the list 4. Select every kth element from the list 5. Divide number of population by desired sample size

Simple Random Sample (Probability Sampling)

1. Obtain a list of all population members 2. Assigns number to all members 3. Randomly select numbers until desired size is reached (random number generator)

4 characteristics of a field study

1. Particularistic- focuses on a particular instance of a problem 2. Descriptive- in-depth description of issue 3. Understanding- Helps understanding of an issue 4. Inductive

Three common moderator problems

1. Personal bias- looking for a specific answer/pushing the convo where it needs to go 2. Unconscious need to please client- same as personal bias 3. Need for consistency- Ignoring things that don't fit

Problems reducing coding reliability

1. Poor definition of categories 2. Unclear instruction given to coders 3. Unanticipated content that is difficult to code

3 Research Paradigms

1. Positivism (quantitative)- Uses tools to measure reality and test assumptions using science 2. Interpretive (qualitative)- Understand how individuals interpret meanings & events 3. Critical (qualitative)- examines power structures and their influence on media and society

Things to be aware of when conducting interview

1. Power dynamics- don't want a hierarchy 2. Judging people 3. Talking too much 4. Not paying attention 5. Coming off as cold

What are the seven factors that determine the sample size required for a study?

1. Project Type: focus groups call for smaller samples, studies call for larger 2. Project Purpose 3. Project Complexity 4. Amount of Error Tolerated 5. Time Constraints 6. Financial Constraints 7. Previous Research in the Area

Four benefits that hypothesis offer researchers

1. Provide direction for a study 2. Eliminate trial and error research 3. Rule out intervening and confounding variable 4. Allow for quantification variables

Characteristics of the Scientific Method

1. Public- available for others to scrutinize 2. Objective- rules and procedures must be followed, free of bias 3. Empirical- can be tested & measured 4. Systematic & cumulative- Research should be built on previous work, nothing is ever "proven" with science 5. Science is predictive- Helps predict future 6. Self-Correcting- Bad results/ideas will eventuallly be rejected

Multi-Stage Cluster Sampling

1. Randomly select "clusters" 2. Randomly select participants from within the clusters

3 Assumptions of Qualitative Data

1. Reality is socially constructed there is no reality 2. Each person's experience is unique and cannot be generalized 3. Value is in deep understanding of a situation, rather than being able to apply it to other situations

Research Process Steps

1. Select Topic 2. Review theory/literature 3. Develop Question 4. Design study- qualitative or quantitative 5. Collect Data 6. Analyze and interpret 8. Inform others

6 things to take notes of during field observation

1. Setting-place/layout 2. Participants- gender, attire, etc. 3. Dialog- quotes 4. Your own feelings 5. Your own actions 6. What you don't understand

3 interview alternatives

1. Solicited narratives- ask them to write about the topic 2. Log-interview- log what they do in a day 3. Personal documents- diaries, letters, etc.

Measures of Reliability (1)

1. Stability: consistency of measure at different points in time e.x. (test-retest, measure the same thing on different people)

What are the five general problems associated with survey research?

1. Subjects are unable to recall information about themselves or their activities 2. Subjects lack feeling or knowledge on a topic 3. Subjects may deceive researchers by giving false responses 4. Subjects given far too elaborate responses 5. Surveys can be too complex for subjects to explain their true feelings.

Four types of leadership behavior/styles

1. Supportive leader- shows concern for well-being of the group, friendly. Best type 2. Directive leader- tell people what to do. Guides/directs. Sets up performance targets 3. Participative leader- consult with group members about activities, asks for suggestions 4. Achievement-Orientated leader- set challenging goals, seeks improvements. Holds members to high standards

8 Problems with developing questions

1. Too broad 2. Only about communications 3. Underlying assumptions/values driving the ? 4. More than one question at a time 5. Already have an answer 6. Too focused on researchers values/interpretations 7. What-if questions 8. Should questions

Survey Design

1. Use limited number of question styles 2. Group Questions together using a similar style 3. Sensitive Questions should go last

Four checks for qualitative data

1. Use of multiple data (triangulation) 2. Audit Trail- protocol, coding descriptions, notes 3. Check with members- Ask participants if descriptions were accurate 4. Check with research team- see if everyone is on the same page

What are the five advantages of survey methods?

1. Used to investigate problems in realistic settings 2. The cost of surveys are reasonable (gathers a lot of information) 3. Large amount of data can be collected with ease. 4. Surveys are not constrained by geographic boundaries 5. Helpful data already exists (data archive, government records, etc.)

4 phases encouraging mass media research

1. WW1 (propaganda) 2. Realization by advertisers that research data can be used to develop ways to persuade customers 3. Increasing interest in the effects of media on the public 4. Increased competition among media for advertising $$

Two sources of variance that ANOVA answers

1. Within group variance 2. Between group variance

What are the four issues a researcher should consider when making a decision about whether to use a probability or non-probability sample?

1.Purpose of the study: some research studies are designed to investigate the relationship between variables, not generalize the results to a population. Nonprobability sampling is the best for examining relationships. 2.Cost vs. Value: A sample should produce the greatest value for the least investment. Probability sampling may cost too much for the type and quality of information it provides. 3.Time constraints: probability sampling may cause researchers to have conflict with time constraints. 4.Amount of Acceptable Error: in studies where error control is not a prime concern, nonprobability sampling is usually okay.

Important Moments in research ethics

1930's-1945- Nazi Experiments 1930's-1970- Tuskagee Experiments 1940's- Guatemala Syphillis Experiment 1963- Milgram Study 1971- Standford Prision Experiment

Measures of Reliability (2)

2. Internal consistency: measure assigns similar values e.x. Split Half Test: First 5 questions get similar scores as the second 5 questions

Measures of Reliability (3)

3. Equivalency: Develop two versions of a scale (different measure of the same concept) e.x. Intercoder reliability: see if results can be reproduced by others

Dependent variable (DV)

A dependent variable is what you measure in the experiment and what is affected during the experiment. The dependent variable responds to the independent variable. -Effect -Outcome

What question is crosstab/chisquare good at answering?

Is A related to B? Does that differ from an expected value? Need two nominal variables

Composite Week

A superior technique to both a random sample and a constructive sample when dealing with newspaper content

Frequency Distribution

A table of scores ordered according to magnitude and frequence of the occurence

What is a dummy variable

A variable measured as the nominal level may be used in higher-order statistics if it is converted into another form. Results of the process are dummy variables

How many papers get retracted from journals?

About 500 or .004% but more are incorrect

Paradigm

Accepted set of theories, procedures, and assumptions about how researchers look at the world.

Online Survey

Advantages: Cheapest, quickest, no interviewer influence, wide range of responses, can have audio-visual content Disadvantages: Hard to get a representative, big brother concerns, must be self-explanatory, length Unique steps: Program survey, follow up

Face to Face Survey

Advantages: Clarification possible, can have audio-visual content, reliable high response rate, length Disadvantages: Interviewer influence, Expensive, Lots of time. Unique Steps: Same as telephone

Mail Survey

Advantages: Cost, length, no interviewer influence, can have visual content Disadvantages: Must be self-explanatory, low response rate (1-5%), collection takes LOTS of time How to increase mail survey participation: Advance mailings, follow-up mailings, University logo, Personal touches (real stamps, signatures) Offer compensation. Unique Steps: Write cover letter, create the package, distribute survey, monitor rates, follow up

Telephone Survey

Advantages: Quick data collection, moderate cost, ability to clarify questions, moderate response rate (once you get someone to answer, 60% non-response) Disadvantage: Some interviewer influence, short questionnaire, no visual depictions. Unique steps: Prepare interviewer instructions, train interviewer, make callbacks, verify data, call up and verify

Intercoder Reliability

Agreement between coders

Intracoder Reliability

Agreement for same coder

Z-score/Standard Score

Allow researchers to compare scores or measurements obtained from totally different methods; allow for comparisons of "apples and oranges"

Continuos Measurement

Allows for subparts (Interval, Ratio)

What is research?

An attempt to discover something

Cut & Paste Technique

Analyzing a focus group transcript- decide what is relevant and code. Coded copy can be cut apart and sorted.

Skip Pattern

Answer questions a certain way leads you to other questions based on your response

Surveys

Anytime you systematically ask people about their attitudes, emotions, beliefs, knowledge, intentions, and behaviors in order to: 1. Identify attitudes or behaviors 2. Examine relationships between variable measured

Standard Deviation

Average dispersion expressed in the same units you used

Variance

Average dispersion, measures average distance in data from mean.

Levels of Measure (1)

Nominal: Variable is measured with categories. -Numbers are meaningless -Must be mutually exclusive (no overlap) -Must be exhaustive (everything fits in a category) e.x. Political Party

Visual Notetaker

Notes psychical reactions at least two

Two pieces of material needed for focus group

Consent form & protocol

Median

Number where 50% is above and %50 below, must have at least ordinal

Histogram

Bar Chart, where vertical bars represent frequencies

Internal validity

Control over research situation to rule out alternative explanations- knowing your study is correct

One-group pre-test post-test design

O1 X O2 Threat to validity: Maturation

Difference between pre-experimental design and a true experiment

Pre-experimental design has 1. No randomization 2. No control group

Focus group transcript

Complete record of verbal comments, including ums and ahs. Typically does not need to include body language

Skewness

Concentration of scores around a particular point/value at the low or high end of the scale -Data can be screwed right/positive or left/negative

Types of Survey Questions (2)

Close-ended: participants choose from responses

Solomon Four Group Design

Combines all designs. 1 group tested before, manipulated, tested again 1 group tested before, not manipulated, tested again 1 group not tested before, manipulated, tested again 1 group not tested before, not manipulated, tested again Threat to validity: random assignment, sensitization

What is correlation test good at answering?

How is A related to B?

Sampling

How you select your subjects

Probing question

Designed to elicit specific information/follow up question e.x. Can you provide an example

Total Sum/Sum of Squared Errors (SSE)

Deviation compared to national average was a negative number To deal with negative numbers, sqaure deviance.

Conceptual (constitutive) Definition

Dictionary definition- using words to describe concept

Deviation

Difference between an observation for that variable

Range

Difference between the smallest and largest observations

Discrete Measurement

Does not allow for subparts (nominal)

Verbal Notetaker

Helps with transcription, Identify when people speak, note quotes that speak out, note general conclusions. Two or more

What is a T-Test good at answering?

Is A greater than B? Can compare means between two different questions e.x. are people closer to parents or coworkers

Population

Entire group of people you want to study-important to clearly define

Emergent coding

Established categories after a preliminary examination of the data. Resulting category system constructed based on common factors/themes

Priori Coding

Establishes categories before data is collected, either on theoretical or conceptual rationale.

Factorial Design

Experimental studies with two or more independent variables -Important to understand whether the combination of two variables increases effects 2x2 factorial design

External validity

Extent to which you can say your results will be found across populations, settings, and times Can be generalized

Major advantage of field experiment

External Validity- study conditions closely resemble natural settings, provides a truer picture of their normal behavior

Confounding Variable

Extraneous variables that might contaminate their findings can be controlled through environmental, experimental manipulations, experimental design, or assignement of subjects.

Type 11 Error (False Negative)

Fail to reject when it's actually false -not finding a difference that is there

Conceptual fit

Fit between operational & conceptual definition e.x. Conceptual Definition: Violence Operational Definition: How many times a character is hit, beaten, or killed

Focus Group

Focused group interview, discussion is recorded and transcribed Analyze for similar themes & ideas

Structured interview

Follows a certain timeline/instructions

Deductive Thinking

General to specific (positivism) theory-hypothesis-observation-confirmation

What is informed consent

Giving participant an overview of the study so they can decide if they want to participate, with consent people know they're in a study

Where should the moderator sit?

In front, reinforcing leadership

Non-equivalent control group design

In the nonequivalent control group design, a treatment group and a comparison group are compared using pretest and posttest measures. Threat to validity: Sampling bias

Static Group Design

In the static group comparison study, two groups are chosen, one of which receives the treatment and the other does not. A posttest score is then determined to measure the difference, after treatment, between the two groups. Threat to validity: Sampling Bias

Volunteer Sampling (non-probability sampling)

Individuals volunteer to be included

Reactivity

Influence that a subjects awareness of being measured/observed has on behavior

Levels of Measure (3)

Interval: Measurement points with equal distance between them -assume scales are equal distance apart -no true zero e.x. strongly disagree-strongly agree

Difference between lab experiment/field experiment

Lab Experiment: High Internal Validity, Low External Field Experiment: Low Internal Validity, High External

Moderator

Leads focus group, keeps group on track, encourage participation/discourage groupthink, keep members comfortable, don't differentiate themselves from group

Two Types of Focus Group Experts

Legitimate experts-can inhibit discussion still Self-Appointed experts- more difficult, seldom have genuine expertise. May intimidate other group members

How can we informally observe a correlation?

Looking at a scatter plot

Inferential Statistics

Looks at two or more variables at the same time. Allows us to make statistical inferences about the relationship between variables

Bimodality

Lots of respondents at the end of each scale, not people are outliers. In this case, report mode

SD & Margin of Error

Margin of error is two SD away from the mean

Reliability

Measure that provides consistent results

Anova Test

Measures means across categories of a nominal variable e.x. do men and women differ in levels of self-esteem Sex is IV gender is DV

Reliable Measure

Measures the same thing every time (consistent)

Valid Measure

Measures what is says it measures (correct)

What could be the single most important factor in focus group success/failure

Moderator

Interrupted Time Series Design

Multiple Pre/Post test observations Design can only have one group

Correlation Test

Need 2 Quantitative Variables How is A related to B Positive: Individuals with higher X have higher Y Negative: Individuals with higher X have lower Y

Field Experiments

Observe subjects in their natural environment

Post-Test Only

Only test group after manipulation Threat to validity- no way to asses if these groups were different to begin with

Types of Survey Questions (1)

Open-ended: participants generate responses

Five types of main questions

Opening- easy, factual question to break ice Introduction- general impressions of the topic Transition- guide participants towards topic Key- 2-5 addressing heart of the issue Ending- summarize what has been said, ask for clarification

Neutral

Opposite of loaded, more simple

Issues with repeated measure design

Order effects: solved with counterbalanced design Fatigue: People get sick of the experiment, people get worse because they don't want to do it again

Levels of Measure (2)

Ordinal: Measure with rank-order categories, tells you where people are placed but no how far apart e.x. gold, silver, bronze

Social Desirability Bias

Over-reporting desirable favors and underreporting undesirable behaviors

Non-Response Error

People might not participant

What test will you use in an ANOVA if you have more than two variables?

Post-Hoc: Test all groups against each other

Where should the notetakers sit?

Present but inconspicuous

Loaded/Direct Question

Pressure or force participants to answer in a particular manner. Choose one answer or the other

Push Polls

Pretends to be a phone survey, real goal is to persuade. Statements are false or bend truth

At Risk Populations

Prisoners, Kids, Mentally Incapacitated Anyone that the researcher has unfair advantage over

Operational definition

Procedures to measure a concept

Pros/Cons of field studies

Pros: Provides detail about situation, good at figuring out why something occurred Negatives: Very time consuming, problems with qualitative data- lack of generalizability & scientific rigor

Power Analysis

Provides an estimate of the minimum number of respondents needs to provide the best chance to discover if something does not exist

Difference between quasi-experimental and true experiment

Quasi-experimental has no random assignment

Random Assignment

Random assignments of participants, each participant has equal chance of getting either group. Creates similar groups in different conditions

Non-Sampling Error (1)

Random error: Survey questions give you different results across samples -very hard to detect, source of error unknown

Random Digit Dialing (Probability Sampling)

Random sampling. You start with a given area code, obtain all the given exchanges in the area code . Start with a number and add a predetermined number

Levels of Measure (4)

Ratio: Measurement points with equal distance between them

Scale

Represents a composite measure of a variable, it is based on more than one item

Quota Sampling (non-probability sampling)

Research samples with a certain number of subjects in categories but does not use random sampling methods

Measurement

Researcher assigns numerals to objects, events, or properties according to certain rules.

What is an unqualified volunteer sample?

Researchers have no control over the respondents or subjects who participate in a research study - almost anyone can participate. Have become uncommon

What does the Institutional Review Board examine?

Reviews/Determines 1. Harm to participants 2. Benefits to Society 3. Protecting the privacy of respondents Private studies don't go through this process

Small SD

Scores for people tend to be close together

Large SD

Scores for people tend to be spread out

Steps outlined in the protocol

Script for a moderator -welcome -introduction -consent form -questions -conclusion/thanks

Universal Survey Steps

Select a sample, construct the questionnaire, collect data, analyze data/calculate responses

What is a qualified volunteer sample?

Selected using probability sampling and limits the type of person who is invited to participate

Y-axis

Shows frequency of the scores

X-axis

Shows the scores, listed from lowest to highest value

Ockham's Razor

Simplest approach is always the best

Method

Specific technique for gathering information following the assumptions of the chosen methodology

Inductive Thinking

Specific to general observation-pattern-hypothesis-theory

Funnel Approach to Questioning

Starting with general questions and moving to more specific

Hypothesis Question

Statement of testable, expected relationship between two topics

Null Hypothesis

States there is no relationship between two variables. Start with the assumption that there is no relationship and find evidenence to prove otherwise.

Margin of Error

Statistical estimate of what the population looks like (How close your results are to actual population)

Descriptive Statistics

Summarizing the data

Non-Sampling Error (2)

Systematic error: results are wrong, but always in the same direction. Easier to figure out

Content Analysis

Systematically examining the content of communication used to 1. Describe how much/what messages exist 2. Compare media content to the "real world" 3. Examine content trends over time 4. Get a starting point for effects research

Mean

Take everyone's score on a variable and divide by number of participants (must have ratio or interval data)

Pearsons R

Tells us how strong a correlation is .3-.5=weak .5-.7=moderate over .7= strong relationship

Cross-Tab Test

Tests a relationship between two nominal variables Is a related to b Will be useful for coding (e.x. does content vary by source)

Pre-Post Test

Tests participants both before and after manipulation. Can "check" random assignment Provides info on change Threat to validity- sensitization

Confidence Level

The certainty that your results fall in a given margin of error

Location&Participation

The closer the location is to the participants, the more likely they are to participate

Homogenity/Compatibility

The more compatible the group members, the greater the interaction and the more open the communication

Operationalization

The process of translating conceptual definitions into something that can be measured

Regions of rejections

The proportion of the area in which the null hypothesis is rejected

Snowball Sampling (non-probability sampling)

The researcher approaches one subject, asks that subject to suggest others, and those subjects suggest others

Mode

The response that occurs most often Can get mode with any level of measurement

Outliers

Throws off data, if there are extreme outliers median is best to report.

Problems with Four Group Design

Time consuming, requires a lot of participants

Latin Square Design

Used for repeated measurements experiment with four subjects. Each subject is exposed to all conditions and that each four conditions appears only once per row & column

Control Variable

Used to ensure that results of the study are due to independent variable and not another source

Representative/Probability

Used to generalize to a larger population

Stratified Random Sampling (Probability Sampling)

Used to get representation from a certain group 1. Divide population into groups that are important to study 2. Randomly sample from each group until you reach desired sample

Non-representative/non-probability

Used when you are looking for a specific characteristic

Repeated Measure Design

Uses same subjects with every treatment, every person serves as their own control

What is ANOVA good at answering?

What are the differences between groups?

Alternative Hypothesis

What the researcher thinks the relationship will look like 1. Directional hypothesis (one-way/one-tailed) Violent game play will cause more aggression than non-violent game play 2. Non-directional- There will be differences between violent and non violent game play on aggression

Connivence Sampling (non-probability sampling)

When subjects are selected based on availability to the researcher, not the purpose of the project

Census

When the entire population is measured

Research Question

When the researcher is unsure of the relationship between two concepts

Isomorphism

Whether the measure corresponds with an actual number (reality)

Type 1 Error (False Positive)

Wrongly rejecting null hypothesis when it's actually true -finding a difference that isn't there -determined by sig level

What is Deception?

You may need to deceive participants for study to be effective If you use deception, you must debrief participants about the nature of the study once they've finished

Sampling Error

Your sample will never be a perfect representation of the population because of chance

Independent variable (IV)

a variable (often denoted by x ) whose variation does not depend on that of another. -Cause -Predictor

Secondary Question

can be open/closed and are designed to follow up primary questions or probe in greater detail

Primary Question

designed to help introduce new topics/areas within a topic

What does a null hypothesis mean for ANOVA

every variable = the same

Case study

examines a single "system of action"-total emersion Uses multiple techniques to examine the case -Field observation -Textual analysis -Focus group/interviewing

P Value

i.e. Significance Value- Most common sig val used is (p < .05) If its less than .05 you can reject the null

Noise

i.e. error- things that could potentially interfere with your research

In-depth

less structured, probing and open-ended questions

Leadership

process of influencing from activities, goal setting and goal achievement

Methodology

study of methods and underpinning philosophical assumptions of the research process itself

One-Shot Case Study

subjects are presented with some type of treatment, such as a semester of college work experience, and then the outcome measure is applied, such as college grades. Like all experimental designs, the goal is to determine if the treatment had any effect on the outcome.

Hypothesis Testing

tells us if we reject or fail to reject the null hypothesis

Manipulation Check

test to determine whether the manipulation of the independent variable that actually has the intended effect

Unit of Analysis

the smallest element of a content analysis, but also one of the most important In written content- a single word/symbol, theme, or an entire article In TV/Film- characters acts or entire programs or films

Counterbalanced design

use separate groups of subjects, each group receiving treatments in a different order.

What would be the number of 'breakoffs' that would suggest the questionnaire is too long?

• Breakoff: Respondent hangs up during a telephone survey or terminates the survey in some way • If a survey has 10% breakoffs, it is too long

What is a response rate?

• Percentage of respondents who complete the questionnaire among those who are contacted/selected

What is a screener/filter question?

• Used to eliminate unwanted respondents; include only respondents who have specific, desirable characteristics for the research

What is weighting or sample balancing?

• When the subject totals in given categories do not reach the necessary population percentages, subject' responses are multiplied (weighted) to allow for the shortfall. • Used to reach predetermined required level.