RESEARCH BRAINTEASER
There are a myriad of over-the-counter, non-prescription cold medicines available to consumers whenever the cold bug bites. However, your client has created the latest and greatest cold remedy. Your task is to conduct research for this new non-prescription cold medicine. You know the first thing on your research To Do list is to develop a sampling plan. (Hint: refer to Chapter 10 if you cannot remember.)
QUESTION
When developing your sampling plan, what is the first step?
When developing your sampling plan, what is the first step?
determine your population
ReplyDeletedefining the population of interest.
ReplyDeleteExactly - figure out who the population for the particular research project is.
ReplyDeleteWho would be the population for this particular research project?
People who have or get colds seasonaly
ReplyDeleteOkay, Correne, but that's a HUGE population. Could you be even more specific? Reread the problem...
ReplyDeleteit would seem like everyone would, but not everyone uses over the counter medicine. so the target would be people who do use it by analyzing the competitor brands and who uses what.
ReplyDeleteFrankie - very good! A researcher may take the position that everyone suffers from colds; therefore, everyone is the population. BUT not everyone buys non-prescription cold medicine when they get a cold.
ReplyDeleteAs you said, we would determine whether people have purchased or used competing brands. Those who have purchased competing brands would be included in population.
People who buy prescription cold relief medicine would be excluded from the population and not included in the research sample.
People who currently have a cold or people who frequently buy over the counter cold medicines
ReplyDeleteMuch closer, Correne.
ReplyDeleteThis week, we will start talking about issues surrounding determining a sample size for our research project (Chapter 11 in the textbook).
ReplyDeleteSince we’ve had a week off, let’s do a quick review before getting into new material!
There are two main methods of selecting a sample method. There’s Probability Samples and Non-probability Samples. What is the difference between the two?
Probability- every member of pop has random chance of being selected
ReplyDeleteNonprob- Non random pop selection
probability - random. everybody could be selected.
ReplyDeletenon probability - not random. eliminating selected people
Great memories! Let's look at Probability Samples first. As you both said, probability samples are those in which every element of population has a random likelihood of selection. The researcher obtains information from a cross-section of the population.
ReplyDeleteAs you may recall, there are four types of probability samples. What are these four?
1. simple random sample
ReplyDelete2. system sample
3. stratified samp
4. cluster samp
simple random sample - random chance of selection
ReplyDeletesystematic sample - population is numbered and you use skip intervals.
stratified sample - every member of population is assigned to a subset and random samples from each subset.
cluster sample - selected from a numer of geographic area
Wow, I'm impressed.
ReplyDeleteGive me an example of when a stratified sample would be used.
Sample people who drive Harleys than divide into women, owners of Harleys, riders only, people who have purchased recently. than take separate samples from all groups.
ReplyDeletewhen you have gathered your population sample and want to go deeper into answering your problem so you go into segments of that population and might even do a sub segment of that in order to find what you need.
ReplyDeletetooth paste for example. people tend to brush their teeth but not everyone has plak problems or wish to taste like mint so there are variety of different brand to include all peoples.
Two good examples. Stratified Sample is when every member of population is assigned to one subset; draw random samples from each subset.
ReplyDeleteEx: Conducting political poll and research shows that there is a difference between how men and women vote; stratified sample is gender.
Okay, let's talk about non-probability samples, which as you both told me, is when members of the population have been selected in a non-random manner.
ReplyDeletePopulation members are selected on basis of convenience, i.e. easy or inexpensive to reach.
There are four types of non-probability samples. Do you recall what these four are?
1. Convenience sample
ReplyDelete2. Judgment sample
3. Quota sample
4. Snow ball sample
Great, Correne. Can you give me short descriptions of each?
ReplyDeleteconvenience sample - use people who are accessible. here and now situation.
ReplyDeletejudgement sample - interviewer decides sample
quota sample - reaching sample sized and being done for the day. a set number reached
snowball sample - asking respondent for referrals.
Excellent Frankie. Must be a combination of great note-taking and memory!
ReplyDelete1. use people who are accessible who is around you i.e. office co workers
ReplyDelete2. interviewer chooses people
3. goal is to reach a certain number of respondents
4. getting referrals from a respondent to continue research
What would be the con/negative to using a Snowball Sample?
ReplyDeleteYou may not get accurate or enough referrals then your sampling snowball would end
ReplyDeletepeople who refer other people might think they would want to participate and in actuality, they do not. also, you have to track these people down that have similar interests by word of mouth.
ReplyDeleteYou both make good points. Another item to keep in mind when using a Snowball Sample is that it your data may result in bias due to lack of true cross-representation from your population.
ReplyDeleteLet’s start talking about how we go about determining a sample size (Chapter 11). There is a general rule to keep in mind for determining sample size. Do you remember what this rule would be? (We've talked about it before. It has to do with errors.)
ReplyDeleteMake it large to avoid sampling errors
ReplyDeleteditto, i agree ^_^
ReplyDeleteRight. The larger the sample size, the smaller the sampling error.
ReplyDeleteNow, on the flipside, there are times when a SMALL sample size is justified. There are three determinants for when a small sample may be used. Can you think of any criteria that justifies using a small (vs large) sample size?
ReplyDeleteWhen budget is a factor
ReplyDeleteYes, that is the first criteria. Money and limited resources (i.e. interviewers) available to you for your research project.
ReplyDeleteThe other two determinants are a little more tough.
ReplyDelete#2: Determine sample size based on what’s been done in the past for other or similar research projects(i.e. sample size of 300).
#3. How many subgroups will be used and is the subgroup large enough to answer the research problem? Ex: If overall sample is 400, but you need 50% women, 50% men, then each subgroup is 200.
rule of thumb and number of subgroups to be analyzed to determine sample size?
ReplyDeleteYou got it Frankie. The "rule of thumb" esp. for larger research projects is that the sample should have 100+ respondents in each major subgroup with 20-50 respondents is less important subgroups. (For your semester research project, I don't expect such large numbers, esp. if you decide to use subgroups!)
ReplyDeleteLet's do a quick review of some statistical terms as this will help you better understand the material to be covered in class on Thursday and in your textbook readings.
ReplyDeleteWhat is a RESPONSE?
Any guesses?!
ReplyDeleteRefers to quantity of response
ReplyDeletehow people answer a certain question or whom their influencers are in answering the question.
ReplyDeleteClose. When we sample, the units that we sample - usually people - supply us with one or more responses. In this sense, a response is a specific measurement value that a sampling unit supplies.
ReplyDeleteWhen we look at the responses that we get for our entire sample, we use a statistic. There are a wide variety of statistics we can use - mean, median, mode, and so on.
ReplyDeleteFor our purposes, what is a MEAN?
Like I said Quantity?
ReplyDeletethe largest number of selected people. a high number
ReplyDeleteThe MEAN is properly computed only from interval or ration metric data.
ReplyDeleteIf you mean in terms of a #, then that is fundamentally correct!
ReplyDeleteThe MEAM is the average you would get. To compute the mean, you add up all the individual values and divide by the total number of values.
For example, the mean or average quiz score is determined by summing all the scores and dividing by the number of students taking the exam.
Consider the test score values:
15, 20, 21, 20, 36, 15, 25, 15
The sum of these 8 values is 167, so the mean is 167/8 = 20.875.
Last one, what is STANDARD DEVIATION? (This is probably a blast-from-the-past of your high school math days!)
ReplyDeletethat does not ring a bell. what math level is that again?
ReplyDeleteThis is how the variance is calculated. The sum of the squared deviations of the mean is divided by the number of observations minus 1.
ReplyDeleteThis comment has been removed by the author.
ReplyDeleteLet's move over to Ethics. We'll talk more about this in class on Thursday!
ReplyDeleteHOMEWORK REMINDER
- Chapter 10 chapter questions 2, 4, 6, 7, 9 due Thursday, March 26th
- Chapter 10 case study with questions #1-3 due Thursday, March 26th
- Read Chapter 11 for Thursday, March 26th
- Part III of semester research project due Thursday, March 26th
Gave you wrong answer for standard deviation - sorry!
ReplyDeleteOkay, that's a hard one. The standard deviation of a sample tells us something about how different samples would be distributed. A standard deviation is the spread of the scores around the average in a single sample.
Again lets take the set of scores:
15,20,21,20,36,15,25,15
To compute the standard deviation, we first find the distance between each value and the mean. We know from above that the mean is 20.875.
So, the differences from the mean are:
15 - 20.875 = -5.875
20 - 20.875 = -0.875
21 - 20.875 = +0.125
20 - 20.875 = -0.875
36 - 20.875 = 15.125
15 - 20.875 = -5.875
25 - 20.875 = +4.125
15 - 20.875 = -5.875