| Term | Definition |
|---|
| Naturalistic Research | Studied in environment with little disturbance by researcher |
| Qualitative Research | Preserve and analyze the situated form, content, and experience of social action rather than subject it to mathematical transformations |
| Interpretative Research | many truths, subjective, free will, natural observation, etc. |
| Ethnographic Research | Researcher control in a natural environment - interviews, police work... |
| Action or Applied Research | Conducted for the purpose of solving a particular 'real-world' socially relevant problem |
| True Experiments | Manipulation, treatment and control groups, and random assignment to groups |
| Quasi-experimental designs | Manipulation, treatment and control groups, NO random assignment |
| Pre-experimental designs | Generally, no manipulation, treatment or control groups, or random assignment |
| Manipulation check | example: manipulation is hurt, manipulation check would be to ask participants how hurtful the conversation was. |
| Decisions to make before experiment | pilot testing? what to tell participants the study is about? may use deception? use confederate? manipulation check? debriefing? |
| Interaction Research | Quantitative method used to make inferences about the effects of manipulating verbal and nonverbal communication behaviors in dyads or groups |
| How to use interactional research methods | example: have participants respond to a hypothetical scenario, role play (decision making), observe conversations |
| How to manipulate casual variables | use a confederate who is trained to behave in a variety of ways, train partners to communicate in different ways |
| Sample | Term used to refer to those people or units that a researcher actually includes in the study |
| Sampling | the process of selecting a set of subjects for a study from a larger population |
| Population | everyone in the area of interest, a universal set - i.e. college students |
| Study Population | The people who fit the sampling frame that could actually be sampled - i.e. PSU college students |
| Sampling Frame | a list of the population from which the sample will be drawn |
| Observation Unit | The unit of data being used- student/individual |
| Sampling Error | Random difference that exists between the sample and the population (If you had the entire population then the sampling error would be zero) |
| Probability Sampling | A sample will be representative of the population from which it is selected if all members of the population have an equal chance of being selected in the sample |
| Random Selection | such that every member of the population has an equal chance of being selected |
| Central Limits Theorem | (mean of scores obtained in population will equal mean of scores obtained in probability sample |
| Random Sample | references the selection of study participants from the population |
| Random Assignment | references how researchers assign participants from the population |
| Internal Validity | the degree that we are successful in eliminating confounding variables within the study itself |
| External Validity | A study that readily allows its findings to generalize to the population at large has high ______ |
| Simple Random Sample | - table of random numbers - computer generated list 1.) Get list of accessible population 2.) Identify size 3.) Use table of random numbers |
| Pros and Cons of Simple Random Sample | Pros: ability to calculate statistical error - Cons: Need for accurate lists, time involved |
| Systematic Sampling with Random Start | the first person is at random and then every Nth person is chosen |
| Stratified Sampling | Organizes population into groups and then randomly samples from subsets (age groups, gender) |
| Cluster Sampling | Identify naturally occurring clusters of participants who have a variable in common with the target population (i.e. divide country into regions: west, southwest, midwest, east, etc., cluster sample randomly selects region to be used in study, once the cluster/region is randomly selected, random sampling within it is used to ensure a probability sample is obtained |
| Pros and Cons of Stratified Random Sample | Pros: ensures various subgroups included - Cons: difficulty in identifying non overlapping groups and the additional time needed |
| Pros and Cons of Cluster Samples | Pros: Save money and time - Cons: is it representative? sampling error |
| Non-probability Sampling | Non-randomly selected participants, data may not represent the population of interest, makes generalization more difficult/less valid |
| Convenience Samples | Based on participant availability |
| Volunteer Samples | Based on participants volunteering |
| Purposive Samples | Based on a predetermined criteria |
| Quota Samples | Nonrandom stratified sampling |
| Network Samples | When participants help recruit other participants; "snowball" sample, can also be purposive - end up with people who are all alike |
| Probability Samples | Simple Random Sample, Stratified Random Sample, Cluster Sample, Systematic Sample, Random Sampling |
| Non-probability Samples | convenience sample, volunteer sample, purposive samples, quota sample, network sample |
| Null hypothesis | predicting that there are no differences or relationships between variables |
| Alternative hypothesis | predicting that there are differences or relationships between variables |
| Identify Null and Alternative hypothesis: H1: Individuals in geographically close relationships are more satisfied than individuals in long distance relationships | Null = there is no difference in satisfaction between the GCRs and the LDRs - Alt: there is a difference in satisfaction between GCRs and LDRs |
| Confidence interval/ level | expressed as a percentage; tells us how certain we can be that the results are real rather than random (generally use a 95% confidence level) |
| Significance testing | determines if a relationship or difference occurred by chance; tests if the null hypothesis is accurate |
| 95% confidence level = | 5% chance the results are based on chance |
| Probability level = | .05 (p-value) allows you to decide if the null is true or false |
| If the p-value is less than .05...... | we reject the null, and accept the alternative - we are 95% certain that the differences are not based on chance |
| If the p-value is greater than .05..... | we accept the null and conclude that there are no differences or relationships between variables |
| Type I error | reject the null when the null should be accepted (say there are no differences when there aren't) |
| Type II error | accept the null when the null should be rejected (say there are no differences, when there are) |
| increase confidence level | To reduce Type I error.... |
| increase sample size or decrease the confidence level | To reduce Type II error.... |
| Chi-square | both variables are categorical, tells us if the differences between groups is statistically significant |
| T-test | IV is categorical with only two categories and DV is continuous (i.e. Do men and women (IV) differ in aggression (DV)? |
| ANOVA | IV is categorical and has three or more categories, DV is continuous (i.e. Do freshman, sophomores, juniors, and seniors (IV) differ in intelligence (DV)? |
| Correlation | How changes in one continuous variable are associated with changes in another continuous variable |
| Correlation Coefficient (r) | strength of the relationship between variables; -1 to 1 |
| Regression | Does the IV predict changes in the DV? (Communication apprehension (IV) predicts changes in heart rate (DV) in response to an impromptu speech |
| Ethics | conforming to the standards of conduct of a given profession or group (morality, right vs. wrong) |
| Informed Consent | research involves no more than minimal risk to participants, if the waiver of informed consent will no adversely affect the rights or welfare of the participants, and if the research could not be carried out if informed consent were required |
| Voluntary Participation & Freedom from Coercion | Researchers cannot offer excessive or inappropriate financial or other inducements to obtain research participants, When research participation is course requirement or opportunity for extra credit, the prospective participant must be given the choice of equitable alternative activities |
| No harm to participants | Minimal risk - no greater than the probability and severity than that ordinarily encountered in daily life or during the performance of routine physical or psychological examinations or tests |
| Confidentiality | the treatment of information an individual has disclosed in a relationship of trust with the expectation that it will not, without permission, be divulged to others in ways inconsistent with the understanding of the original disclosure |
| Anonymity | When a researcher does not know who participated in a study or which results belong to which participants |
| IRB (institutional review board) | approves or denies conducting proposed research based on the ethical principals |
| informed consents, use of information, privacy issues, methods, compensation, costs and benefits of participation | The IRB approves the following... |
| Qualitative Research | studies people communicating in their own environments, guided by their own objectives, and how they give meaning to their communication |
| The 5 commonly shared characteristics of ALL qualitative research | 1. natural settings 2. researcher as participant, 3. subject-based communication, 4. subject intentionality, 5. pragmatism |
| Natural Setting | location, time, rituals are all determined by subjects; enviroment is not and was never designed for experiment |
| Researcher as participant | researcher perceived by subject as a participant in some significant way |
| Subject-Based Communication | subjects allowed to identify and determine topics of communication, provide transactions...not generated by researcher |
| Subject intentionality | researcher seeks to capture and preserve the communication and symbol-using of subjects as they understand and intend them |
| Pragmatism | the research analysis resolves an exisiting social problem. it may or may not contribute to theory development |
| Non-probability / Network | Sally is completing an interview study on homosexual relationships. She is having a hard time accessing the gay community in state college. to increase the number of people in her study, she asks current participants if they will tell other people they know in the homosexual relationships about the study. |
| Probability / systematic | Jen is conducting a study on attitudes of people who live in Pittsburgh. B/c of limited funds, she uses the PGH phonebook as her sampling frame. She begins by opening the book to a page in the the middle of the book and chooses 1 person to begin recruiting. She then calls every 5th person in the phonebook. |
| Non-probability / volunteer or Convenience | Tom is studying drug use on college campuses. He post info about his study in the PSU student listserv to recruit participants |
| Non-probability / purposive or quota | Amy is interested in differences b/w long-distance and geographically close dating relationships. B/c she wants to compare the 2 groups, she recruits 100 individuals in long-distance relationships and 100 individuals in geographically close relationships. |
| naturalistic, qualitative, interpretive, ethnographic, field, and action/applied | What are the 6 different labels that have been used to describe different dimensions or characteristics of qualitative research? |
| Grounded Theory | theory emerges inductively from the data. from the ground up. |
| Reactive Effects of qualitative research | guinea pig effects, response sets, interview effects, change in research instrument, population restrictions and stability |
| Triangulation | a researcher will attempt to combine different research findings into one coherent explanation of human communication |
Combine with other sets
Share on Facebook
Share on MySpace