Wednesday, February 28
Good job with standard deviation today. This is a tough concept and not that much fun but it sounds like your grasping the material. This week in class, we discussed;
Designing questions for surveys
Types of samples
a.) simple random
b.) systematic sample
c.) simple stratified sample
d.) proportional stratified sample
Sampling from periodicity
Descriptive v. Inferential statistics
a.) mean
b.) median
c.) mode
Range/ Variance/ Standard Deviation
Normal Distribution
Standard scores
Frequency tables/ histograms
Last week, we talked about two different types of studies, Longitudinal and Cross sectional. Think of a study where you would need to conduct a longitudinal study. For posting, describe the study and explain why a cross-sectional design would not work. Also, which type of longitudinal design would you use and why (Trend, Cohort, Panel)?
Designing questions for surveys
Types of samples
a.) simple random
b.) systematic sample
c.) simple stratified sample
d.) proportional stratified sample
Sampling from periodicity
Descriptive v. Inferential statistics
a.) mean
b.) median
c.) mode
Range/ Variance/ Standard Deviation
Normal Distribution
Standard scores
Frequency tables/ histograms
Last week, we talked about two different types of studies, Longitudinal and Cross sectional. Think of a study where you would need to conduct a longitudinal study. For posting, describe the study and explain why a cross-sectional design would not work. Also, which type of longitudinal design would you use and why (Trend, Cohort, Panel)?
posted by Dr. Jeffrey Bolt at 5:03 PM
11 Comments:
If you were conducting a study on people's most common choices of fast food restaraunts, you would want to go the longitudinal route. If you were to ask people one day to report the fast food places they go to the most, you surely would not get consistent results over time. People change their eating habits and styles constantly, so a cross sectional study would most likely not yield valid results. Factors such as environment, health/dieting issues, location, and financial issues would all play a role in people's fast-food eating habits. I believe that a panel study would be the best technique to use when conducting this survey because it utilizes the same people over a period of time to learn how their attitudes change.
Jim Sherlock
Another example of a longitudinal survey would be a panel study on the participation of students in an online course. Because a cross-sectional survey would only measure the sample once, it would not be the best way to determine changes in student attitude and/or behavior. The group of registered students could be surveyed pre-course, then one week into the course, followed by mid-course and at the conclusion of the course to measure the changes in their beliefs, attitudes and behaviors toward online coursework.
Marianne Riggenbach
They say that people's political views change over time, so it would be a good topic for a longitudinal study. It would have to be a panel study, so the same people's opinions could be taken throughout the years. This way, when changes in political views take place could be pinpointed. The first time to survey would be between ages 18 and 20, when an individual is first able to vote; the the survey could be taken every 5-8 years following. It could be considered complete when the person reached about age 50 or 60. A cross-sectional study would not work because people's opinions change as they get older and life has been lived.
Amanda Gabriele
You could coduct a study on people's choice of music. When we are young we start off listening to one thing but as we grow older our taste in music change. This would be a panal study so you can have the same people opinions through out the study. This can show when the person change of music started. You can start the study at the age of 12 and go to the age of 30. The survey can be taken every 2 years. A cross-sectional wouldn't work because it will only measure the sample once.
Jeff McConnell
A longitudinal study could be cobducted about how many classes a student missses a semester. this would be a panel study. a cross-sectional study would not be accurate over time because the amount of classes one misses changed each semester. things like how hard or important the class is, as well as if there is an attendence policy would effect the number of classes attended, or missed.
Dan Caparso
If you were conducting a study about a certain weight-loss program using a longitudinal study would be useful. I would use a panel of people and interview them over a long period of time to test the effectiveness of the weight-loss program. Using a cross-sectional study wouldn't work because you only collect data once, and that wouldn't be an good measure of how effective a weight-loss plan was.
Lauren Birmingham
If people wanted to test the side effects of a new medicine they would need to do a longitudinal study. They would need to do this because side effects can happen the next day or 10 years down the road. This way, they can track the effects throughout the years. I believe this would be a cohort study because they are all exposed to the same medicine. The fact that they are taking the same medicine may make them a subgroup? I guess I'm not really sure about that though.
Elizabeth Franko
Another example of a longitudinal survey would be a panel study on people's favorite television show. We enjoy different televsion shows as we grow up and as our taste in televsion changes. We would use a panel study because if we use a panel then we would have the same people on the panel. We could survey people from age 10-50. This would give us a wide survey and would cover most people and the most different types of shows. We could test each person every 5-6 years. People change their televsion viewing habits so much that a cross sectional study wouldn't work.
Richard Toth
One effective way for assessing people’s attitudes towards media preferences is the cohort study. If we are to evaluate college student’s tendency for the use of computer, for example, we might start our cohort research by selecting a specific group of people who have one thing in common; they are all college students who have graduated from American schools between the year 1990 and 2000. We might conduct several surveys at several points in time by asking the same category of people and compare their attitudes to the one’s held by students who have enrolled in colleges from the year 2000 on.
Mohamed Aboulhaya
A study that would have to be longitudinal would be if you were to want to see how people lived their lives depending on how they were raised. You would start when they were relatively younger and question them about school and home activities and some questions about their families. Then maybe once every year or two years you would change the questions to reflect how they have grown up. So eventually you would add college and work questions as well. This would be a panel study so that you can follow the same people over time. This would not work as a cross sectional because people change too frequently.
Machelle Montgomery
Measuring the grade point average of a couple college students as they go through college would be a longitudinal study. This is becasue their grade point average would change as they got older and classes got older.
Justin Quinlan
Post a Comment
<< Home