UNIT 3 FAQ SHORT Flashcards
A 4 year high school of 2000 students, sampling 40 high students: Describe a simple random sample
Number students 1-2000. Use rantom number generator to get 40 unique integers from 1 to 2000.
A 4 year high school of 2000 students, sampling 40 high students: Describe a stratified sample
Stratify by year. Randomly choose 10 FR, 10 SO, 10 JU and 10 SENIORS
A 4 year high school of 2000 students, sampling 40 high students: Describe a convenience sample
Ask the first 40 students coming to the locker rooms after school. This is problematic because athletes may not have the same preferences as non athletes.
A 4 year high school of 2000 students, sampling 40 high students : Describe a systematic sample.
2000/40=50. You need 1 out of each 50.
Get an alphabetical list of all of the students. Randomly choose one of the first 50 students and then every 50th student after that.
A 4 year high school of 2000 students, sampling 40 high students: Describe a cluster sample
Imagine that all of art classes have 10 students and they are mixed with fr, so, jr and srs… You would randomly choose 4 classes and survey everyone in each of the 4 classes.
What is the standard sampling method?
A Simple Random Sample (SRS) is our standard. Every possible group of n individuals has an equal chance of being our sample. That’s what makes it simple. Put the names in a hat.
give an Example of a MULTISTAGE sample
Suppose you want to poll urban, suburban and rural citizens, you can divide a map into those strata, and then randomly choose neighborhoods or streets in each and ask everyone on those streets. Here you stratified by community type and then clustered by street.
What are the two types of observational studies?
Retrospective, and Prospective
What is cluster sampling?
Cluster- grab clusters of the population. Go to a random class and survey everyone, or go to random street and survey everyone.
What is retrospective study?
A retrospective study is a study that looks backwards in time (or at the present moment).
What is systematic sampling?
collecting data from every nth subject.
What is prospective study?
Prospective study is when you study the experimental unit’s present and future.
What is a representative sample?
A sample that looks like the population. It has similar characteristics.
What is stratified sampling?
When you break the population into groups with similar attributes and randomly select from each strata.
What are the “good” sampling methods?
SRS (simple random sample), stratified, clustered, systematic, multistage
What are the “bad” sampling methods.
convenience samples, voluntary samples
What is a weakness of SRS?
If you SRS a high school, you could randomly get ALL FRESHMEN.. which wouldn’t be representative. So you would stratify to make sure you get some from each grade.
Give four examples of an SRS of size 20 from a population of 650.
- Number subjects 1-650 and use RANDINT(1,650) and use first unique 20.
- Number subjects 001-650 and use a random number table. Take 3 digits at a time. Ignore 000 and 651-999. Use the first unique 20.
- Write names on 650 identical index cards, shuffle them. Choose 20.
- Write names on 650 identical tennis balls, put them in a cement mixer for 7 hours. Dive in blindfolded and toss 20 out to your friend. Use those 20.
What is wording bias
A type of response bias, When the wording of the question impacts response to it. (type of response bias). Often language that invokes emotions. For example. “Do you want to support starving children in America?” vs “Do you support giving welfare money to people who don’t work ?”
What is BIAS in sampling?
A systematic FLAW in your method. Under coverage, Wording, Voluntary, Convenience, Comfort (psychological), Response, Non-response BIAS. Even with a larger sample, you will still have bias.
What is a sampling frame?
It is the frame from which you get your sample. For instance, if you call people the frame would be “people with phones,” if FOX news takes a poll, the sampling frame is “fox news watchers”
When sampling, what kind of sample are we striving to get?
A representative sample, we want our sample to have similar charactaristics as the population
To make a survey to tell of a restaurant is good, would you ask the people coming out of the restaurant?
If you had to, what might you add?
People at the restaurant are probably there because they already like it. If you asked the question “Is this your first time dining here?” and if they say “yes” you survey them, that would be a better method. But then again.. the people wouldn’t go into an Italian restaurant if they didn’t like that type of food.
What is undercoverage?
Undercoverage is when a group of the population is not represented in the sample (didn’t have a chance). When the sampling frame isn’t representative.
What is response bias? How do you avoid it?
Response bias is any influence that may sway the respondent e.g wording of the question, interviewer’s behavior/background. Therefore, in a survey, ask questions that allow respondents to answer comfortably and honestly. Keep the wording “indifferent” or neutral in some way in order to unduly favor one response over another.
Things that cause nonresponse bias ?
(remember non response is that the people you ask, or try to ask don’t respond) Lazy researcher, shy survey takers, who is the questioner, environment, person might not want to be judged.
What is statistically significant?
When an observed difference is too odd for us to believe that it is likely to have occurred naturally (or just randomly). Basically it is Statistically Significant when we don’t think it happened randomly. when you think “something’s up” or “something’s fishy”
How can you use random numbers to sample?
Number the subjects 00-99 (if less than 100) or 000-999 (if less than 1000) or 0000 to 9999 etc.. then use a random number table taking one, two, three or four numbers at a time. Throw out repeats.
Name types of bias
undercoverage, non response, response, voluntary
How is BIAS different from SAMPLING error
Bias is a systematic flaw in your sampling method. Sampling error is always present even with the best methodology- it is the natural variablility of sample statistics. Different samples give different statistics.
name some differences between experiments and observational studies:
(can you remember two of these?)
Here are five.
1. Experiments can prove causation (studies can’t) . 2. In experiments, you assign treatments (studies you just watch) 3. In experiments you can use volunteers (in studies you want representative samples). 4. In experiments you make inference about treatment (In studies you make inference about population). 5. In experiments you randomly assign treatments (in studies you randomly choose subjects).
What is the difference between response bias and nonresponse bias?
Response is when the person’s response is influenced by the question or questioning method (like if a parent asks if you use drugs, as opposed to a friend… there is only one answer to this, but one might respond differently to them), non response is is when the people who don’t respond might have different opinions/views than the people who did.
Example of response bias
A teenager goes to the doctor’s office with their parents. The doctor asks the teen if they vape. The teen may say “no” because their parent’s are there, even though they do vape.
What is the main purpose of a placebo ?
To blind the subject that is being experimented on to avoid influence to the given variable therefore altering the response variable . When people think they’re getting help, they often improve anyway..
Example of undercoverage
You only ask people who go to Home Depot about their views on school lunches.
You only ask people in the library about school lunches.
(you only cover people who go to these places)
Example of nonresponse bias
In a survey, a person does not answer a few questions (or a person is on your list and you can’t get a hold of them).
You go to someone’s home and they aren’t there.
You tried to ask.. but got no response.
How can you decrease sampling error?
Get a larger sample
Is sampling error a mistake?
IT IS NOT A MISTAKE!!! Because the data in samples are generally different, the statistics calculated from one sample to another vary and are generally not equal to the parameter. This variablilty of STATISTICS is called sampling error.
Can you eliminate sampling error?
Only if you take a census.
Larger samples tend to have less error, but you can never get rid of sampling error when you take a sample.
Will larger samples reduce BIAS?
No, bias is a systematic flaw, even large samples will still have bias. If you ask more people outside of McDonalds, you still only get answers from people who eat at McDonalds (large samples can reduce error, however)
How can the WORDING of the question lead to response bias
Words or phrases that impact your feelings tend to influence responses. Look for “devastating, horrific, wonderful etc.” Sometimes there is a background story like “Many americans lose jobs to illegal aliens every year, do you feel this is fair”
Obs Study vs. Experiment- when do you manipulate environment?
In an observational study, you just collect data observe things. In an experiment you are manipulating the environment.
Obs study vs Experiment.What are you making inferences about?
In studies, you are making inferences about the population, in an experiment you are making and inference about a treatment.
Obs Study vs. Experiment: How do you use random numbers?
In studies, you are randomly selecting a sample from a population. In a study you are randomly assigning treatments to subjects or subjects to treatments.
Obs Study vs. Experiment: Causation?
You can only prove causation in an Experiment. In a study you can just find an association or a correlation.
4 INGREDIENTS TO EXPERIMENTS
Compare, control , randomization, replication (and BLOCKING when you need to)
4 ingredients: What is “control?”
You want to control the environment as best as you can so that the only difference between groups is the treatment, and the treatment only. Everything else should be similar.
4 ingredients: What is “replication?”
Having enough subjects. You don’t want to test fertilizer on just one plant. You want a few plants, or subjects, in each treatment group.
4 ingredients: What is “compare?”
Having something to compare your treatment with helps you see its effectiveness. This is why we usually have control groups.
4 ingredients: What is “BLOCKING?”
If you think different groups of subjects may respond differently to treatments because of location, gender, age, then you make BLOCKS, and make sure that each block gets all of the treatments - compare within each block. (which med worked best for females?).
What are the (three) types of experiments?
- Completely randomized
- Randomized block
- Matched pairs, but matched pair is blocking with 2 in each block.
What is a control group?
The group that doesn’t get a treatment (or gets the old treatment). It helps us see the impact of the environment. It gets the placebo or standard care but goes through all of the motions
Factors and Levels:
Give medications example
Factor: medication.
Levels: 50mg, 100mg and 200mg.
Factors and Levels:
give “sleep” example
Example. For the Factor “SLEEP” the, level(s) would be how many hours the subjects were alowed to sleep.. 4 hours, 6 hours, 8 hours.. 3 levels
4 ingredients: What is “randomization?”
You want to randomly assign subjects to treatment groups. (or treatments to subjects)
Factors vs Levels:
give example with diets.
DIET PLAN would be a factor and levels could be: low carb, low fat, and no diet
Give example of a confounding variable
If you are doing a fertilizer experiment on cloned tomatoes, and fertilizer A is on left, fertilizer B is on right side of room. You find that fert B has larger tomatoes, but then notice they were closer to the windows/sunlight. Now sunlight is a confounding factor- was it the sunlight or was it the fertilizer?
A weight loss plan has three diets (A. low carb B. low fat C. no change) and three exercise plans (1. walking 2. weights, 3. no change). What are the factors? The levels? many treatments are there? Can you name the treatments?
Factors: Diet and Exercise
3 levels of DIET: A, B C
3 levels of EXERCISE: 1, 2, 3
9 treatments:
A1, A2, A3
B1, B2, B3
C1, C2, C3
Give and example of when you may stratify
If there are different types of subjects in the population, you may want to make sure you get each type. For instance, you may be looking for average tree diameter, so you want to be sure to get some of each tree type in the location. Or if you are sampling students, you may want to make sure you get males and females, so you would break the population up and randomly choose some of each.
How is blocking different from stratifying?
Blocking breaks up subjects into pre-existing groups and then you randomly assign the subjects from each group to all of the treatments. This helps you reduce confounding variables.
Stratifying break the population up in to groups, then you randomly sample from each group- this helps you get a more representative sample.
Give examples of when you may block
If you are studying fertilizer, you may block by plant type, making sure each veggie gets both fertilizers. Or you may block by proximity to sunlight, a “close to window” group and a “far from window” group. For migraine medicine, you may block according to gender, so that both males and females get each of the treatments. You block when you think underlying characteristics might impact the response.
What is a response variable in an experiment?
WHAT YOU ARE MEASURING/comparing!
-mean weight loss
-increased percent participation
-change in mean anxiety score
What is wrong with using volunteers in an experiment?
Not much. In an experiment, we are not looking for a sample that is like the population. We just want to see the effectiveness of a treatment. It is fine if the subjects are all similar. In fact it is best sometimes when they are!
