Flashcards in Part1 Deck (45)
What is sampling?
Sampling is the process of selecting units of analysis (people,cities, records) to study.
What is a: population parameter? Sample statistic? Population? Sample element?
A) Population Parameter: the true statistic we never know; we estimate it based on a sample B) Sample Statistic: the actual statistic we obtain based on our sample C) Population: The larger grouping of study elements we want to generalize our research findings to (high school students, “probationers”, Judges) D) Sample element: The unit about which information is collected (people, cities, etc)
What are conscious versus unconscious biases with regard to sampling?
Conscious Bias: When a researcher purposefully samples individuals that do not reflect the population of interest. Unconscious Bias: When researchers sample individuals that do not reflect the population of interest, but they do not do this on purpose.
What is sampling error?
? Sampling error: How much our sample statistics might be “off” from the real population parameters.
What is standard error? What makes standard error larger or smaller?
Standard Error: The statistical calculation of sampling error. It is based on 2 things: the estimated population parameters( the sampling statistics) and the sample size. S=√P*Q/n. The sample size; increasing the sample size decreases the standard error.
What are confidence levels? Confidence Intervals? (Be sure to know how to calculate a confidence interval for a given confidence level when the sample statistic and standard error are provided)
Confidence Levels: The estimated probability that a population parameter will be within a given interval. Confidence Interval: The estimated range of values that include the population parameters Based on confidence levels, sample statistics, standard error. Refer to notes
How does the level of confidence we want to speak to influence the size of a confidence interval?
The confidence interval for a 95% confidence level will be twice as large as the interval for a 68% confidence level.
What is the difference between a population and a sampling frame? What implications does this difference have for a study?
Sampling frame: the available sample elements within a population from which a sample is drawn.
What is probability sampling?
Probability Sampling: A method of selecting sampling elements where the probability that an element will be selected is known. It minimizes biases and produces valid standard errors
What are the 4 different types of Probability Sampling Designs?
A) Simple Random Sampling: Each elements in the sampling frame is assigned a number, and numbers are randomly selected for inclusion. B) Systematic Sampling: Randomly sampling every C) Stratified Sampling: SRS sampling within lists based on key sample element characteristics. D) Multistage Cluster Sampling: Random sampling at different stages (Levels) when a complete list of sample elements is not readily available.
What does stratification attempt to achieve?
The purpose is to reduce sampling error for variables that are stratified.
What type of sampling designs do the NCVS and BCS employ?
The NCVS and BCS use the Multistage cluster Sampling. Refer to page 243 bottom paragraph
What is non-probability sampling?
A method of selecting sampling elements where the probability that an element will be selected may not be known. Ex are purposive, quota, and snowball sampling
What are the 4 different types of Non-Probability sampling Designs? What are the distinguishing features of each of these designs?
A) Purposive Sampling: Sample elements are eligible for selection based on the specific purpose of the study at hand. In other words selection is not based on a random process as I is with probability sampling strategies. B) Quota Sampling: Recruiting subjects for a study until you have met specific quotas (usually based on sex, race, age). The sample looks representative of the population, but selection is not based on a random process. C) Reliance on available subjects sampling: Subjects are recruited based on simply accessibility. The key question is whether people who are accessible are also typical of the larger population. D) Snowball Sampling: Starting with a single participant who then refers you to other similar individuals. Common strategy with “hidden” populations in field research settings.
What is surveying?
Surveying is the method of collecting data by using a standardized instrument. It is the most frequently used data collection method in the social sciences.
What is an open-ended question? Closed-ended?
Open-ended question: Respondent provides own answer. Benefits of OEQ: Respondent gets to speak in his/her own terms; may provide answer that the researcher had not anticipated. Shortcomings of OEQ: Researcher must code and categorize each response; potential for respondent to skip over the question. Closed-ended question: Respondent selects answer from list.
What are the different principles of K.I.S.S.?
“ Keep It Simple Stupid”. The easier it is for someone to process a statement/ question the more likely they will respond. A) Use plain language B) Shorter items are better C) Avoid double barreled questions D) Avoid negatively worded items E) Avoid prestige bias: the tendency for respondents to agree or disagree with an issue because a prominent person or organizations opinion is revealed. Ex. American Idol judges opinion
What are the 4 different ways in which we can try to reduce error associated with memory recall?
A) Bounded Recall: respondents answer a question based on the time period since they were last questioned. B) Narrow the preference period: Data will be more related if respondents are asked to consider shorter periods of time. C) Averaging: Having respondents answer items based on typical patterns of behavior. D) Land marking: Basing reference periods on a major event or holiday in the past. Ex. Start of school year, since your birthday, since the beginning of 2013.
What is social desirability bias? How can we try and minimize it?
Social desirability bias is the tendency of respondents to answer questions in a manner that will be viewed favorably by others. It can take the form of over-reporting "good behavior" or under-reporting "bad," or undesirable behavior. The tendency poses a serious problem with conducting research with self-reports, especially questionnaires. This bias interferes with the interpretation of average tendencies as well as individual differences. Topics where socially desirable responding (SDR) is of special concern are self-reports of abilities, personality, sexual behavior, and drug use. When confronted with the question "How often do you masturbate?", for example, respondents may be pressured by the societal taboo against masturbation, and either under-report the frequency or avoid answering the question. Therefore the mean rates of masturbation derived from self-report surveys are likely to be severe underestimates. We can minimize it by anonymous administration is preferably used as the person does not feel directly and personally involved in the answers he or she is going to give. And administer test through computer.
What are contingency questions? Matrix design questions?
Contingency Questions: Follow up questions that only some respondents will answer based on prior answer. Matrix Design Questions: a block of related questions with the same response categories for respondents to choose from.
What relevance does question ordering have for survey research?
Beginning questions should be sterile and non-threating but interesting to respondents. Very sensitive and personal questions should come later in the survey. Ordering may differ according to the mode of administration.
What are the 3 different forms of Self-Administered Surveys?
Self-Administer Surveys: Respondents fill out questionaries’ their own. A) Mail surveys B) Electronics Based survey C) Group based survey
What are the major things to consider/do when choosing to use a Mail Survey method?
“Heads up” mailing- postcards; cover letters- established legitacimy; include return envelopes; and follow-up mailings
What are response rates? How are they calculated?
Response Rates: number of individuals who completed and returned the survey divided by the sample size (75% good; 70% excellent)
What are the different types of electronic surveys? What limitations are there when using electronic surveys?
Electronic based survey (email; website) Potential issues could be spam filters, easily diverted; representative sample.
What is In-Person Survey Interviewing?
In- Person Interviewing: researcher is present with respondent and reads questions/ statements.
What is probing? Why is it important?
Probing: asking respondents to clarify or be more descriptive with answers. Probe when needed- especially with open-ended questions.
What is CAI? CASI? ACASI?
A) Computer Assisted Interviewing (CAI): the researcher enters responses straight into a computer database (e.g. Excel) B) Computer Assisted Self-interviewing (CASI): the respondent reads questions and enters responses into computer. C) Audio and computer assisted self interviews (ACASI): Respondents listens, reads, and answers questions on a computer.
What are the benefits and disadvantages of CAI?
Benefits: Quick transfer of responses into electronic format; Complex skip patterns can be built into the questionnaire; Can program indifferent languages; Audio supplements for those who cannot read; keeps respondents on a task. Disadvantages: High cost of computer/ programming; Errors in programing can be problematic; Technology issues- what if computer crashes