ADPR 3850 1st Edition Lecture 7 Outline of Last Lecture I Introduction to Research Outline of Current Lecture II Sampling Probability v Nonprobability III Sampling Error and Sample Size IV History behind Research V Opinion leaders VI Induction v Deduction VII The Research process VIII 8 elements of a program plan Current Lecture Today Super Bowl PR Continue with Research Planning PR the Super Bowl JCPenny tweets drunkenly three times about the game THEN JCPenny tweets they were tweeting with mittens to explain the typos publicity stunt or cover up o Professor s thought is that the company isn t doing well wants to bring attention followers to JCPenny Sampling Probability vs Nonprobability sampling Probability Sampling each element of the population has a o Non zero o Know and o Equal chance Of being selected into the sample Nonprobability Sampling o One of the assumptions of probability sampling is violated e g TV call in polls there are certain types of people more likely to be viewing it internet surveys If poll goes out at 2 pm fathers are at work mothers may be home Important in order to calculate sampling error i e a measure of how precise a poll really is we need to use probability sampling techniques These notes represent a detailed interpretation of the professor s lecture GradeBuddy is best used as a supplement to your own notes not as a substitute Types of Probability Sampling Simple Random Sampling SRS o Every element and every combination of elements has an equal chance of being selected o Problem requires a list of sampling units o requires access to telephone maybe everyone has the opportunity Systematic Sampling o Choose a starting point and every nth unit is chosen to be selected o ex Phonebook call every 10th person o Problem requires a list of sampling units Cluster Sampling o Divide population into clusters and then randomly choose X number of clusters may also sample within clusters Types of Nonprobability Sampling Accidental Haphazard Convenience Sampling o Exit polls o Call in polls Purposive Judgemental Sampling o typical precincts o Medical research o you don t want random sample ex medical wants to find treatment for asthma so you will want to test those with asthma seeking out people based on certain characteristic Snowball Sampling o Invitation based on your sample o involves sending a link to fb friends maybe and asking them to pass it on to people in their social network grows Quota Sampling o Once most common sampling technique o Involves sampling according to pre defined characteristics of population o we know x of population is Hispanic so seek out that percent Samples Sampling and Population Distributions If you ask 5 or so you don t know what you will get could be 5 democrats or 5 republicans no idea But if you ask 1 000 you will get a solid idea of where people are actually going to vote The more you ask the closer you get to reality your margin of error shrinks you get much more accurate results a normal distribution o This is what you really need to know about sampling Sampling Error and Sample Size Diminishing Returns At some point you will reach diminishing returns So say we are calling in election data you want to know if democratic or republican candidate is more likely to win A graph shows margin of error and sample size If we only ask 200 people there is a higher margin of error The more people you ask the lower the margin of error Increasing sample size decreases margin of error But at some point you are bringing in all these people and it costs money to sample more and it doesn t do much for margin of error From 600 to 2 000 people you are decreasing the margin of error by maybe 1 but you can stop because you will hit a point of diminishing returns So for this reason 1 000 is usually the stopping point The Development of Opinion and Market Research Before 1940s o Na ve outlook regarding research and what it could do based on WWI Propaganda Orson Welles War of the Worlds radiobroadcast causing widespread panic thought alien invasion o Lack of commonly accepted research procedures o Magic Bullet media as magic bullet we blindly follow During and immediately after WWII o Involvement of government in applied research o Columbia s Bureau of Applied Social Research Lazarsfeld et al wrote The People s Choice New developments in research techniques We started to understanding that sampling size matters and how we ask survey questions matters Reinforcement and two step flow we moved from idea of media as magic to idea that media is about reinforcement and two step flow so if I hear something I already believe in I will accept it as the idea is reinforced Two step flow is the media influences key opinion leaders in opinion leader then you follow them People that matter to you are doing it During and after 1950s o Sharp increase in privately funded PR and advertising research o Emergence of telephone and online polling o As a result development of industry research as we know it today o Contingent and delayed effects media affects our contingent contingent on things like our values political beliefs meaning media will influence certain people more than others the degree to which we readily accept media information delayed effects sleeper effect meaning sometimes media will tell you something and you ll dismiss it but a week later the residual effects are still there maybe subconsciously familiarity from the ads you saw of something Opinion Leaders Has been around for a long time and will continue to be around Ex Chase Bank and opinion leaders o Chase Bank will ask sorority fraternity members to pass message along and they will give vouchers for restaurants concert ticket opportunities geared toward university they see college students as opinion leaders and give us perks and they will pass along this info through their social networks If you hit a few key important individuals it will influence the masses Using Research Ways to use research o Achieve credibility with management I can show you why this will work based on experiment survey literature o Define segment publics What messages will work for them o Formulate strategy Getting sense what they want build strategy accordingly o Test messages Test slogan send it to focus group first do interviews then have evidence to support slogan or not o Prevent crises o Monitor competition PR curious how people respond do their tweets FB so monitor social media o Generate publicity EX JCPenny tweet o