COM 285 Exam 1 Flashcards
Name the three unscientific methods of knowing
Tenacity: It is true because it has always been true, it is true because others think its true
Intuition: I just know it will work, because I know
Authority: It is true because powerful people told me so
What are the characteristics of scientific research?
Public: Everyone can conduct scientific research if properly trained. Even when scientific result is privately owned, its accuracy can be tested using publicly acknowledged methods and standards
Objective: Researchers change, research findings stay the same
Empirical: Research process and results should be indefinable, measurable, testable, knowable
List four qualitative research methods
- Field Observation
- Focus Group
- In-depth Interviews
- Case Studies
Case Study
In depth, detailed examination of a particular case within a real world context.
In-Depth Interviews
Open ended; explore a respondents POV, experiences, and feelings
Focus Group
Group interview with people who share similar characteristics or interests
Field Observation
Observes a targeted person or group in their environment to gain insight on their behavior
List 4 quantitative media research methods
- Content analysis
- Survey
- Longitudinal Research
- Experiment
Content Analysis
Research tool used to determine the presence of certain words, themes, or concepts within some given qualitative data. Using content analysis, researchers can quantify and analyze the presence, meanings, and relationships of such certain words, themes, or concepts.
Survey
Collecting information from a pool of respondents by asking multiple survey questions. This research type includes the recruitment of individuals collection, and analysis of data.
Longitudinal Research
Researchers repeatedly examine the same individuals to detect any changes that might occur over a period of time. Longitudinal studies are a type of correlational research in which researchers observe and collect data on a number of variables without trying to influence those variables.
Experiment
Establishes procedures that allow the researcher to test a hypothesis and to systematically and scientifically study causal relationships among variables.
What is the basic procedure for media research?
1) Select a topic 2) Research on background information 3) Pose the research question and testable hypothesis 4) Demise research methodology and research design 5) Collect data 6) Analyze and interpret the data 7) Present the results
**Certain research questions can only be answered by using certain methods; you can’t choose any method you want
Concept & Construct
An abstract idea generalized or summarized from particular observations
*Concepts describe the empirical world
*concepts that are less observable or less measurable are called constructs
Variable
A counterpart of concept/construct in the empirical world. A concept that can be measured and varies
Qualitative Research Advantages
1) Investigate the issue in its natural setting without artificial manipulation
2) Flexible; allows researchers to investigate the issue as deeply as the research process goes. No limit
3) Allows the researcher to pursuit new research. directions discovered in the research process
Qualitative Research Disadvantages
Data reliability can be low, because research results might be dependent on a small group of participants
Independent Variable
The factors that might influence the dependent variable
Dependent Variable
The concept that the researcher wishes to explain
Levels of quantification/measurement
Binary
Nominal
Ordinal
Interval
Binary Level
yes or no/ true or false response
Nominal Level
Used to label variables without any quantitative value. Ex; select your party affiliation
1 republican
2 democrat
3 independent
Ordinal Level
Used to measure variables in a natural order or ranking. Ex; socioeconomic status, education level etc.
Interval Level
The order and difference between two values is meaningful - Ex; temperate. Disadvantage: no true zero
Reliability
Stability of the research deign. If you redo the study again on the same group of subjects, will you be able to get the exact same results?
Internal Validity
The extent to which the study is designed to investigate the proposed research question. Are you really studying what you want to study?
External Validity
The extent to which the research outcome can be generalized across populations, time, and settings. Is the research finding true in other settings?
Quantitative Research Advantages
High level of reliability
Higher level of external validity
Quantitative Research Disadvantages
Not flexible
IRB HSRB
Institutional Review Board
Human Subject Review Board
Informed Consent
Potential participants must know the general topic of the study before making a decision -> Researchers are obligated to reveal any information that might influence the decision-making process of potential participants -> Researcher must reveal ALL risks of participating in the study to potential participants
Voluntary Participation
No forced decisions
Participants can exit the study whenever they want
Concealment and Deception
Yes, when?
1. Theres no other way to study what you are trying to study
2. Benefit > Harm for subjects
3. Participants can withdraw at anytime without penalty
4. Any potential damage is temporary
5. Researchers MUST debrief the subjects ASAP, and the research procedure is available for public view
Anonymity vs Confidentiality
Respondents cannot possibly be linked to any particular responses
Even though respondents can be identified as individuals, their names will never be publicly associated with the information they provided
Population
ALL the subjects you want to study
Sample
A “subset” of the population or the subjects you recruited to the study
The unit of analysis
Can be one person, one family, one social media account, one organization, etc - population is not limited to a person
Nonprobability Sampling
Self-selected sampling
Snowball Sampling
Nonprobability quota sampling
convenience sampling
Self Selected Sampling
Procedure: Send out the recruiting information to as many potential participants as possible. Those who are interested in the study contact the researchers to join the study.
Snowball Sampling
A researcher randomly contacts a few qualified respondents and then asks these people for the names of relatives, friends, or acquaintances they know who may also qualify for the research study.
Non-Probability Quota Sampling
Procedure: Theres one variable that is extremely important to the study. Based on that variable, divide the sample size into categories. Each category has a specific quota of sample size. Use other Nonprobability sampling method to recruit participant for each categories
Convenience Sampling
Units are selected for inclusion in the sample because they are the easiest for the researcher to access.
Probability Sampling
Simple Random Sampling
Systematic Random Sampling
Stratified Sampling
Clustered Sampling
Simple Random Sampling
Procedure: Number All units in population from 1-100. Use computer program to randomly draw 100 number from the 1000 numbers. Find the correspondent units and you have your sample.
Systematic Random Sampling
Procedure: (Select every nth subject.)
1. Number all units from 1 to 1000
2. Determine the interval n.
Q: do you randomly pick a interval? or you have to calculate an interval?
Interval = population size / sample size
3. Randomly select a starting number from 1 to n.
4. From the starting number, select every nth number, until the end of the population.
5. Find the correspondent units and you have your sample.
Stratified Sampling
Procedure:
1. Divide population into categories A, B, C, and D, based on an important variable.
2. Randomly select certain number of subjects from each category.
Q: How to decide the number of participants within each category?
The numbers of subjects for categories are determined by their proportions in the population.
e.g. 20 from A, 30 from B, 25 from C, 25 from D
3. Combine them all and you have your sample.
Clustered Sampling
Procedure:
e.g. National study
50 states -> randomly select 5 states
From those 5 states -> randomly select 5 cities from each state
Conduct research in the 25 selected cities as a nationally representative study.
Q: If researchers use SRS, instead of CS, what is the sampling procedure?
What should you consider when deciding between using probability methods vs. non-probability methods?
- Purpose of the study
- Cost vs. Value
- Time Constraints
- Amount of acceptable (sampling) error
Research Error
The difference between true population score and the score collected from the sample.
Research Error includes Sampling Error & Measurement Error. Define them.
Sampling error: Error derived from the difference between sample characteristics and population characteristics
Measurement Error: Error derives from how the researcher studies the constructs
Self Selected Sampling Advantages
The recruiting process is normally easier since researchers are sending the same information to many potential participants at a time.
Self Selected Sampling Disadvantages
Self selection bias (those people choose to participate in the study for a reason, that reason might skew the research outcome).
Snowball Sampling Advantages
Very handy in recruiting specific types of participants, who are hard to reach through mass media recruitment or convenience sampling.
Snowball Sampling Disadvantages
People typically bring people who have similar profile of them. Keep getting the similar kind of participants.
Convenience Sampling Advantages
Easiest sampling method
-> low rejection rate
Convenience Sampling Disadvantages
High “interviewer bias”
(participant might feel they are obligated to offer some kind of study result, because they know the researchers.)
Quota Sampling Advantages
Better representation of the population
Quota Sampling Disadvantages
Takes a little more effort than just using other Nonprobability sampling methods
Simple Random Sampling Advantages
Each subject in the population has an absolutely equal chance of being selected. (calculable generalizability)
Simple Random Sampling Disadvantages
-> Sometimes a complete list of population is not attainable.
-> Expensive, time-consuming, and requires a lot effort.
(having the numbers is easy, recruiting the correspondent subjects is hard)
Systematic Random Sampling Advantages
-> Selection process is easier.
(only randomly select once, for the starting number).
(When finding the correspondent participants, it’s easier).
Systematic Random Sampling Disadvantages
-> still need a complete list
-> Periodicity might bias the sample.
Stratified Sampling Advantages
-> Higher external validity (reduced the chance of bias compared with simple random sampling)
-> Maximum representativeness of the stratified variable, which is often important to the research question.
Stratified Sampling Disadvantages
-> Knowledge of the population (regarding the stratified variable) is required prior to sampling.
-> sampling procedure is more time-consuming than SRS
Cluster Sampling Advantages
-> Complete list of all subjects within the population is not needed
-> Clustered subjects are easier to reach and recruit to the study
Cluster Sampling Disadvantages
-> Clusters might not be representative to the population
(To gain convenience, external validity is sacrificed)
-> Each subject must be assignable to a specific cluster
