APR 280 Week 5 Notes
APR 280 Week 5 Notes APR 280
Popular in Investigation and Insights
Popular in Advertising
This 4 page Class Notes was uploaded by Tricia Sylvia on Friday March 25, 2016. The Class Notes belongs to APR 280 at University of Alabama - Tuscaloosa taught by Brandon K. Chicotsky in Spring 2016. Since its upload, it has received 11 views. For similar materials see Investigation and Insights in Advertising at University of Alabama - Tuscaloosa.
Reviews for APR 280 Week 5 Notes
Report this Material
What is Karma?
Karma is the currency of StudySoup.
You can buy or earn more Karma at anytime and redeem it for class notes, study guides, flashcards, and more!
Date Created: 03/25/16
Lesson 13 Thursday, March 3, 2016 8:41 AM Inherent Error in Sampling • This is where formulaic statistics comes in • Practitioners/researchers employ simple random sampling by clearly and unambiguously identifying each member of a population through the use of a comprehensive sampling frame Systematic random sampling 1. Determine final number of interviews needed for a study, or total sample size 2. Determine sampling interval o Divide the number of elements in the sampling frame (total population) by desired total sample size o The result is a number (n) • Generate sample by selecting every nth element from a sampling frame • First sample element must be se lected randomly Periodicity • Bias that occurs when a sampling list has a cyclical repetition of some population characteristic that coincides with a sampling interval Cluster Sampling • Using groups, rather than individuals • Each cluster serves as a sample element • Geographical designations (one neighborhood/one university) • Used when an exhaustive list of elements is not available o Ex: tornado disaster relief Myth: • Large sample sizes are best • WRONG o Representativeness is key Myth: • Researches should sample a fixed % of a population to produce an acceptable sample size • WRONG o Probability based sampling uses mathematical calculations to debunk this Myth: • Researches should bases sample sizes on industry standards or "typical" sample siz es used in other research projects • WRONG o Make thoughtful decisions based on individual requirements of research needs Sample Distribution • A grouping or arrangement of a characteristic that researchers measure for each sample member • It reflects the frequency with which researchers assign sample characteristics to each point on a measurement scale • Common in survey research. • What's measured might include o Opinions o Attitudes o Behaviors o And related characteristics Standard Deviation • The Standard Deviation is a measure of how spread out numbers are. Its symbol is σ (the Greek letter sigma) The formula is easy: it is the square root of the Variance. Lesson 12 Tuesday, March 1, 2016 8:14 AM Parsimony: • If something is parsimonious it is simple-‐-‐easily applicable to other areas of research Model: • Helps explain relationships in a way that is easy to classify Sampling Methods • Has direct relationship with generalizability Probability and Non-‐Probability sampling methods o Probability is more reliable o Probability means everyone has the same chance of getting selected o Non-‐probability is not as random Population • Constitutes all the members of a group or an entire collection of objects • In PR it is the target public Parameters • A characteristic or property of a population, which are found or determined after collecting data o Ex: if a random probability of 1000 students, 35% own Uggs, you can say 35% of all students are Ugg owners • Assessed after data is collected • There is some error, but small enough to infer characteristics of populations with a high degree of confidence o Sample is a subset of a population or universe Representativeness • Would the size of the sample add validity if its not representative of the target demographic o NO! It's not a part of the target demographic Sample frame • List of members in a target audience Sample unit • Individual persons from this list Probability Sampling • Random samples • Produce results that are highly generalizable to a population Non Probability Sampling • Low in generalizability and external validity so why use it? o Quick and easy to generate o Lower cost • Precursor research, or something before a major study (pretest) • Lack of generalizability is a serious limitation o Not selecting randomly, not everyone has an equal chance of being selected 1. Convenience sampling • Ex: college students • Why is sample representation limited? o Because only people exposed to the survey will hav e access to it • E.g. GQ Magazine 2. Quota sampling • To get your quota 3. Dimensional sampling • Making sure that someone from a demographic is represented 4. Purposive (judgmental) sampling • A subjective judgment • Example: only picking "cool" people • Forces diversification-‐-‐not representative 5. Volunteer sampling 6. Snowball sampling • You get a volunteer and ask them to grab someone who will grab someone etc. Stratified sampling • Breaking up the sample into smaller groups and then picking people from each strata (group) Simple Random Sampling • Most scientifically valid methodological Forms of simple random sampling 1. Systematic sampling • Example: picking every seven people, but first person must be random 2. Stratified sampling • Break into groups, select equal samples from e ach group 3. Cluster sampling • Take homogenous groups, break them into groups then proportionally draw from each group • Why simple random sampling o Helps eliminate sample bias o Reduces the chance of subgroup overrepresentation or underrepresentation Inherent Error in Sampling • This is where formulaic statistics comes in
Are you sure you want to buy this material for
You're already Subscribed!
Looks like you've already subscribed to StudySoup, you won't need to purchase another subscription to get this material. To access this material simply click 'View Full Document'