PCQ10_StudentID: 100962404
Nickname: Sunny
Q3: D
Honorcode: 1
Date: 3/29/00
Time: 3:24:42 PM
You can make statements about causation from the results of the study dividing the Freshman into groups which study for different amounts of time, since the subjects in each group are similiar in age and situation. The controlled amount of study hall time can be assumed to cause the difference between the GPAs of the groups. In the first study, although it may be significant, you may not be able to say that the independent variable(class) is the cause of the dependent variable(number of hours studied per week) since there may be other variables that have to do with the students study habits, such as activities.
A Mean Square is the estimated variance. In ANOVA, you can calculate the estimated variation of scores within each of the groups and the estimated variation of scores between groups. In a rare instance, the Mean Squared Between and the Mean Square Within can be equal. One way this would happen would be when there is no variance within groups and between groups. Another unusual way would be if the formulas worked out so that the MS between groups and the MS within groups were calculated to be the same number using the formula SS/df. If the MS within and the MS between were equal, the F ratio would be 1, according to the formula MS between/MS within.
2 - I had trouble with this one because I was unsure of the second part of the question. After not finding an answer in the chapter, I used logic and common sense, although I assume there is probably another answer. 3 - it took me a while to figure out that the formula in the book was referring to. I gathered from looking through the chapter that S sub M is the same thing as the square root of MS total. However, in order to find MS, I would need to know the number of subjects used in the study or the total df. This is why I put answer D, not enough information. One thing that threw me also is that I did not understand what (R2) meant.
PCQ10_StudentID: 101186427
Nickname: Spunk
Q3: D
Honorcode: 1
Date: 3/29/00
Time: 3:56:03 PM
WE can make statements about the causation in study number 2 and not study number 1. In Study 2 the freshman are divided into groups where each group studies different length of study hall time per week. We can compare the means within the groups and bewteen the groups and we know that the means are all coming from the freshman class= Not a third variable. We can compare the GPA to these different length of study hall time. With that, we can see if the amount of study hours does in fact increase GPA. In Study 1 the idependent variable is the class, the the dependent variable is the number of hours studied per week. A third variable may have an effect on the way the subject is studing. They may be studing with the radio or TV or even at a partying - you just don't know. In study 2, you know that the students are studing in a quiet atmosphere where there should be no distraction. Plus- if you are trying to see which class(freshman, sophmore, junior, senior) studies more- other factors may influence that such as, seniors may be sick of studing, freshman tend to focus on general education classes which don't usual require as much studying time as major classes, or seniors and juniors may study more than freshman and sophmores because they are in their major classes. Overall other factors may affect Study 1.
The mean square is another name for the variance because the variance is the mean of the squared deviations. When the null hypothesis is true, the means squared between should be the same as the mean squared within. The F ratio is the ratio of the between group to the within group. If these two groups were the equal then the ratio would be 1. The only time you can reject a Null hypothesis is when the F ratio is larger than 1. So, if the mean squared between is equal to the mean squared within then you can not reject the Null Hypothesis. You have to assume that it is true.
PCQ10_StudentID: 101780234
Nickname: lucky
Q3: D
Honorcode: 1
Date: 3/29/00
Time: 5:29:13 PM
In general, to determine causality, the independent variable (x) and the dependent variable (y) must be associated in some way, x must cause y, and no other factors should precede x that could cause both x and y. In addition, there must be a clear understanding of how x causes y. Therefore, we can only make statements about causation from the results of the second study. In this study, year in school is controlled (the groups are all freshman) so that variable has no affect on x or y. So, x is the only variable left open to cause y; and, x and y are associated (in general, the more one studies, the better he/she does, and the higher his/her semester GPA will be). The first study, on the other hand, only presents a relationship between x and y and does not give a clear understanding of how x caused y or how one's year in college affects the number of hours he/she spends studying per week.
A mean square is a mean of the squared difference between two means either within groups or between groups. In between groups, the formula is the sum of (n (M-GM)^2)/(the number of groups - 1), and in within groups, the formula is the sum of (x-M)^2/(the number of subjects - the number of groups). For the mean square between to be equal to the mean square within, all of the samples in the experiment must have come from populations with identical means, variances, and shapes. Therefore, both mean squares (Between and Within) should be equal or about equal. In turn, the f ratio will be 1 to 1 allowing a researcher to accept the null hypothesis and reject the research hypothesis.
number 3 was hard
PCQ10_StudentID: 101220954
Nickname: Nash
Q3: D
Honorcode: 1
Date: 3/29/00
Time: 7:07:49 PM
The study that takes a group of Freshman and randomly divides them into 4 groups to be assigned a different length study time is the study that we can make a statement about causatoin from. This is because the variation among means taken from the same population being freshman is directly related to the variation of scores within those popluations. In this experiment we should be able to use the variation in the means of our samples to figure out how much variation there is in the populations these samples come from. There is also more control in this experiment because the times are assigned so the data will be in equal ratios and more the results will be more precise.
A mean square is the estimated variance of a distribution of means. In the analysis of variance whether the null hypothesis is to be rejected or not plays a vital role when comparing the Mean Square between and the Mean square within. When you do not reject the null hypothesis, all the populations have the same mean. Any variation among these means of your particular sample thus can only represent variation among individual scores in the population. Therefore, the variance of individual scores can be estimated from variation among samples. When the null hypothesis is rejected, the populations have different means. Any variation among means of your particular samples thus can represent both variation among individual scores in the populations plus variation between the population means. Therefore if the null hypothesis is not rejected then the MS between and MS within will be about the same. This would cause the proportion used to find F (between-group/within-group) to be about 1 when the null hypothesis is true.
PCQ10_StudentID: 100984765
Nickname: Boomer
Q3: B
Honorcode: 1
Date: 3/29/00
Time: 8:05:24 PM
We can make statements about causation regarding Study Two. This studies independent variable or "cause" is the different length study-hall times. This is controlled and therefore much easier to predict the outcome. We will know specifically how many hours each student will be studying. Also, all of the participants are Freshman, which also makes it less diverse than the population of Study One. Study One's independent variable is the academic class or year. This variable is very diverse and concrete stamement could not be drawn simply from their year in school. Unlike the study hours in Study Two, Study One's independent variable is hs many more outside factors. The year of school brings in various influences which makes it difficult to make statements about what the actual cause is.
A mean square "is another name for the variance." If the null hypothesis is true, then the Mean Square Between and The Mean Square Within will be approximately equal. The F ratio is the ratio of these two values, therefore the F ratio would be around 1.
PCQ10_StudentID: 101385622
Nickname: Kit Kat
Q3: D
Honorcode: 1
Date: 3/29/00
Time: 9:08:07 PM
You cannot make statements about causation from the first study because it is a correlational design. There may be many confounding variables such as a more difficult course load for Juniors and Seniors . You can conclude that there is a correltation between year and time spent studying, but you cannot conclude that one causes the other. You also cannot infer a direction of causality from a study that does not use random assignment. The second experiment uses random assignment. Each subject is randomly assigned to a certain length of study-hall time. In this case, one can infer that time spent in study hall has an effect on GPA.
The Mean Square is the population variance as estimated from the sample. The variability between the samples would be the same as the variability within each sample. Since the variability within each sample is attributed to random error, one would conclude that the variability between the samples is also only attributed to random error, and not due to treatment. It is ideal for the variability between the samples to be much higher, showing that it is due to treatment. In such a case, the F ratio would approach one, which is never a significant ratio.
We haven't yet learned how to compute effect size for number three. The book is very unclear on this process.
PCQ10_StudentID: 100936160
Nickname: littlebug
Q3: B
Honorcode: 1
Date: 3/29/00
Time: 11:32:19 PM
In study number two, we can make statements about causation, whereas study number one would show correlation. A students GPA can cause a person to spend more time in study hall; for example, if a student gets a low GPA, that might cause he/she to spend more time in study hall. Vise versa, a students amount of time in study hall could effect he/she's GPA; for example, a student who spends 4 hours a day in study hall, should have a better GPA than one who doesn't spend any time in study hall. As for study number one, the class (freshman, sophomore, junior and senior) would should a correlation with the number of hours spent in study hall.
The "mean squared" is another name for the variance because the variance is the mean of the squared deviations." The Mean Squared Between is the variance of the distribution of means times the sample size, whereas the Mean Squares Within is the average of population variance estimates computed from each sample. "The F ratio is the ratio of the between group estimate of the population variance to the within-group estimate of the population variance." F ratio = Mean Square Between/Mean Square Within. If the Mean Square Between equals the Mean Square Within, the F ratio will equal 1; this means that the experimenter would have to find the Null Hypothesis to appear true, because the 2 population variances would be based on the same thing (the variation within each of the populations), and the division of the overall deviation into two parts should be random.
PCQ10_StudentID: 101736154
Nickname: mercury
Q3: B
Honorcode: 1
Date: 3/30/00
Time: 10:20:32 AM
We can make statements about causation from the results of Study #2. In the first study, freshman, sophomores, juniors, and seniors are being compared as to how many hours they study, but it doesnt show you what causes freshman to study any more or less than juniors, for example. You are comparing groups of people, not what causes something. In the second study, hpwever, you are comparing the number of hours studied and comparing the difference among one group of people and how that reflects GPA. This would show you how the number of hours studied maybe a cause for your GPA.
A mean square is another name for the variance because in the analysis of variance, the variance is the mean of squared deviations. Things that may cause the results of an experiment to make the mean square between equal to the mean square within would be if the had the same, or little diffreence in population variance. If the mean square between was equal to the mean square within, the F ratio would equal one, since F= MS(BG)/MS(WG).
PCQ10_StudentID: 101337519
Nickname: Trixy
Q3: D
Honorcode: 1
Date: 3/30/00
Time: 2:01:24 PM
We can make statements about causation from the results of the second study because these participants were not assigned to experimental groups to ensure similar averages on background variables, they were randomly divided into 4 groups. In the first study the participants were assigned to experimental groups to ensure similar averages on background variables because they were divided by their class year. When participants are assigned to groups to ensure similar averages on background variables the power is normally reduced. This happens because the procedure reduces the contribution of random variance to the between group estimate which is the numerator in the f ratio, but not to the within group estimate which is the denominator in the f ratio. This therefore makes the f ratio smaller and so we will accept the null hypothesis.
A mean square is another name for the variance. It is the sum of squares divided by the degrees of freedom. If the null hypothesis is true then the variability between groups would equal the variablility within groups, which would mean that the treatment would not have an effect. The f ratio would approach a value of one.
PCQ10_StudentID: 100959428
Nickname: butterfly
Q3: B
Honorcode: 1
Date: 3/30/00
Time: 2:03:30 PM
I would say that we can make a statement about causation from the results of the second study because the independent variable(the different length of study-hall time per week) determines the dependent variable(the semester GPA).
A mean square is another name for the variance because the variance is the mean of the squared deviations. When the null hypothesis is true, the within group and the betwwen group estimates are based on the same thing, which might cause the results of an experiment to make MSb be equal to MSw. This would affect the F ratio by making the answer be 1, because you divide MSbetween/MSwithin and if these numbers were the same that would equal 1.
I thought number 1 was quite tricky!!
PCQ10_StudentID: 100962404
Nickname: Sunny
Q3: A
Honorcode: 1
Date: 3/30/00
Time: 3:13:53 PM
see previously submitted answer!
see previously submitted answer!
Thanks for letting me re-submit...I'm still unsure of the answer, but I gave it another shot!
PCQ10_StudentID: 101905648
Nickname: rygy
Q3: B
Honorcode: 1
Date: 3/30/00
Time: 3:53:24 PM
You can make statements of causation about the results from the second study because there is a possibility that the length of study hall each week might affect a student’s GPA. It can’t be the first study because the class of the student won’t affect how many hours a week they study. In the second study the results of how long a students study hall is could have an effect whether positive or negative on student’s GPA.
A mean square is another name for the variance because the variance is the mean of the squared deviations. It is also sometimes called the error variance. A psychologist named Fisher called it Mean Square. In an ANOVA, results that show of an experiment that make the Mean Square Between be equal to the Mean Square Within are. If the Mean of Square Between were to be equal to the Mean of Square within the F ratio would be affected because it would be very close to one. Meaning the chance of rejecting the null hypothesis would be extremely lessened, unless the number of degrees of freedom was extremely high for the within group.
PCQ10_StudentID: 100652169
Nickname: monkeydoodle
Q3: B
Honorcode: 1
Date: 3/30/00
Time: 3:58:24 PM
We could make a statement about causation from the second study because it uses a follow up multiple comparison, post hoc, to examine the subgroup of freshman. This comparison, which is not used in the first study, helps to protect against the possibility of getting significant results just by chance since a great many comparisons can be made.
A mean square is another name for the variance because the variance is the mean of the squared deviations. If the mean square between was equal to the mean square within, then the F ratio would only be 1 and therefore it would be unlikely for us to be able to reject the null hypothesis.
PCQ10_StudentID: 101250465
Nickname: Pebbles
Q3: B
Honorcode: 1
Date: 3/30/00
Time: 4:20:20 PM
We can make statements about causation from the results of the first test: Freshman, Sophomores, Juniors, and Seniors compared to their study time. We can because this is more structured (by groups) than the second which is done randomly by Freshman. The second one mandates study time which means that the student can either utilize that time well or not. Whereas you can get better results when asking how much they study on their own instead of "forced" studying.
A Mean Square is another name for the variance because the variance is the mean of the squared deviations. If the MS between equaled the MS Within then F would equal one because F=MSb/MSw. If SSbetween equaled SSwithin and the degrees of freedom were the same then MS Within would equal MS Between.
PCQ10_StudentID: 101954741
Nickname: angelfire
Q3: B
Honorcode: 1
Date: 3/30/00
Time: 5:10:53 PM
Study two which systematically assigns its participants to experimental groups to ensure similar averages on background variables is going to have a smaller power. This is because the procedure reduces the contribution of random variance to the between-group estimate but not to the within-group estimate. According to Ross and Klein, group matching reduces the F ratio. A smaller F ratio means a smaller chance of getting significance even if there is a true mean difference between the populations represented by the experimental conditions--that is, the power is lowered. But they also point out that this applies on the average and that it is quite possible that in certain specifiable situations, the F ratio might actually be increased by this procedure. However, the traditional advice of textbooks of experimental design is not to use group matching of this sort in setting up experiments.
Mean square is sometimes called the "error variance" and can be defined as the variance, because the variance is the mean of the squared deviations. The f ratio is the ratio of the between-group estimate of the population variance to the within-group estimate of the population variance, and therefore, would equal one if the mean sqaure between was equal to the mean square within. If the F ratio is one, then the null hypothesis is true, and the treatment would NOT have an effect.
PCQ10_StudentID: 101194842
Nickname: moggy
Q3: B
Honorcode: 1
Date: 3/30/00
Time: 5:23:01 PM
You can make a statement about causation from the results of the second study using only freshmen with different lengths of study lounge time. There is no basis for comparison in the first study. The second study will compare the amount of time in the study lounge to their semester GPAs. This way you would be able to see if more time in the study lounge would increase the freshmen's GPA for the semester.
The mean squared is the sum of squares divided by the degrees of freedom. If the null hypothesis is true, then these two estimates of the standard deviation should be the same. This would make the f-ratio about one.
PCQ10_StudentID: 101264596
Nickname: droopy
Q3: D
Honorcode: 1
Date: 3/30/00
Time: 5:29:54 PM
The second study is the one that we can make assumptions about causation between the two variables. The reason is that the first one is not subjected to a level of an independent variable. They are just different groups:freshman, sophomore, junior and senior. The second group is subjected to different levels of an independent variable(study-hall hours per week).
The mean square is the sum of squares(either between groups or within groups) divided by the degrees of freedom. The mean square between groups and the mean square within groups would equal each other when the null hypothesis is true and the treatment would not have an effect. The F ratio would then approach a ratio of 1.0.
I don't believe we'd learned anything about effect size in class yet for ANOVA and the book uses different terminology the last problem caused some trouble.
PCQ10_StudentID: 101847140
Nickname: tigerlily
Q3: D
Honorcode: 1
Date: 3/30/00
Time: 5:51:18 PM
Study One: INDEPENDENT = 4 seperate classes DEPENDENT = #of hours studying / wk. Study Two: INDEPENDENT = different lengths of study halls DEPENDENT = gpa
I believe Study Two would be the case that we could make statements about causation from. This is because this study uses one main group (Freshman) and divides it into four subgroups within a main group. This leads to a Within Group variance. All the students in the second study are from the same intellect level (on a whole, being freshman), and given different hours of study time may cause the gpa's to increase or decrease. You can make statements of causation due to the same class divided into various groups, but these various groups have different study hall hours. Therefore, we could conclude that the different hours affect gpa's of Freshmen in one direction or the other. In the case of the first study, we could not make statements about causation due to many factors. One, the classes are all on different intellectual levels (Freshman having less class experience than Seniors); two, the Seniors may have harder classes than any of the Juniors, Juniors in opposition to Sophomores, or Sophomores in opposition to Freshman; finally three, the number of hours studied per week can still differ given that one class requires more time spent to understand and learn concepts than another. These four, divided groups are completely seperate and would therefore show larger variance between groups.
Mean Square is another name for the variance because the variance is the mean of the square deviations. The formula is MS = SS/df. If the null hypothesis is true then the variability between the groups will be equal to the variability within groups. In essence, the 'treatment' would not have an affect and the F ratio would approach a value of 1.0 .
PCQ10_StudentID: 100734537
Nickname: cheesecake
Q3: B
Honorcode: 1
Date: 3/30/00
Time: 6:38:30 PM
We can make statements about causation for the results from the second study and not from the first. This is because each of the groups comes from the same overall population, which is freshmen college students, who share an average GPA. With the first study, though, the three groups are three entirely different populations, all of which would have their own mean. If each group being tested shares the same population mean, then a difference in the dependent variable will be completely explained by the independent variable.
The Mean Square is equal to the variance (the average of the sum of squares) when doing an ANOVA. If the mean square between is equal to the mean square within, then the F ratio will be 1, and the null hypothesis will not be rejected. This would occur if the populations had the same mean, and therefore the variance between groups would be the same as the variance within a group, and would be due entirely to random error (not treatment)
PCQ10_StudentID: 101115311
Nickname: scoobs
Q3: B
Honorcode: 1
Date: 3/30/00
Time: 6:52:55 PM
We can make statements about causation from the results of study number two. This is because, in this study, the sample is divided into 4 groups randomly and then the independent variable is applied in varying levels.
In the ANOVA, the mean square is another name for the variance. If the mean square between was equal to the mean square within, the variance is the same within the groups as it is between. This might be caused by the null hypothesis being true. The within and between group numbers are estimates of the same population. The F Ratio would then be 1 and we fail to reject the null.
PCQ10_StudentID: 101967921
Nickname: Betty Boop
Q3: D
Honorcode: 1
Date: 3/30/00
Time: 7:40:37 PM
A few different statements about causation could be made concerning these two studies. In study one, the number of hours studied between the classes would vary. I would assume that different factors, such as school involvement, getting ready for college, and "senioritis" would effect how much they study. But in the second study, the indepdent variable is made up of all freshman, students who are in the same category, and their GPA's would be effected by the amount of assigned study time. So, overall, causation in the first study is effected by outside factors, and in the second study those outside factors are controlled.
A Mean Square is another way to describe variance, because variance is the mean of the square deviations. If the mean square between and the mean square within were equal, that would mean that there is little or no variance between the measurements or scores. Also, this would show the null hypothesis were true. The F ratio would be very close or equal to one in this case because it is a ratio of those two numbers.
Question number one was really difficult. I couldn't find anything about it in the text or in my notes. I basically used logic in my answer, and things I knew about causation in general.
PCQ10_StudentID: 101169420
Nickname: quackers
Q3: B
Honorcode: 1
Date: 3/30/00
Time: 8:06:35 PM
I would choose the second group b/c the independent and dependent variables actually depend on each other more. the correlation i believe would come out stronger. the fact that the four groups are coming from the same population also helps to insure that they are more normally distributed.
The square mean is the same as variance, there is both with in group and between group estimates of the population variance. To have the mean square b/w to equal the mean square with in you would have to accept Ho then the variance of the two means would be equal. This means that the treatment would not have an significant effect and the F-ratio would approach the value of one.
PCQ10_StudentID: 101274025
Nickname: SA
Q3: D
Honorcode: 1
Date: 3/30/00
Time: 8:51:52 PM
Statements about causation can made when referring to the 2nd study. In the first study, class year is being correlated to how much one studies. There is such a large variety between the 4 groups that you cannot necessarily say whether or not the amount of studying they do is an effect of age or of other factors. However, in the 2nd study, where freshman are split up and given certain study-hall times, it is more likely that you can make statements about causation. This is because you are singling out studying and GPA. The more one studies, the more likely their GPA will go up. That would be one correlation. Since it is so narrow, it makes more sense to make comments about causation.
The mean square is equal to the variance in the ANOVA tests. In these tests, the assumption of the null hypothesis is that the means on d.v. among the populations are equal. So, if MSbg and MSwg were equal, then that would mean that the null hypothesis was true and that the independent variable did not have an effect on whatever it was that was being measured. The F ratio would reach a value of 1 if MSbg and MSwg were equal.
PCQ10_StudentID: 101968018
Nickname: nicholas
Q3: B
Honorcode: 1
Date: 3/30/00
Time: 9:27:13 PM
You're unable to make any type of statements about the first study because it uses class as an independent variable, however, dependent upon the class, the hours studied per week can vary, making it impossible to make a statement about causation. Whereas in the second study, the students all have the same amount of study-hall time per week which is consistent and allows u to make statements about causation from the results of the dependent variable.
The mean square is in essence a measure of variance, just another name for it. To have the mean square between and the mean square within be equal however, it would mean that the treatmenet would not have an affect and F would approach a value of 1, then the variability between the groups would equal the variability within the groups.
PCQ10_StudentID: 100879061
Nickname: brianimal
Q3: D
Honorcode: 1
Date: 3/30/00
Time: 10:01:13 PM
We can make statements about causation from the results of study two. This is because there is a relationship between the amount of time studying and GPA. If freshman are made to go to study hall for a different amounts of time, it should reflect in their GPA. However, there is no relation between the year in school and the number of hours studied per week. It is not certain that a student from one class level will study the same or a different amount of time than a student from another class level. Th
The Mean Square is the estimated variance based on the variation of the scores with each of the groups. The Mean Square Bewtween and the Mean Square Within could be equal if matching subjects across the groups on variables turn out to be related to the variable studied. This could cause a Type 1 error.
PCQ10_StudentID: 100632956
Nickname: froggy
Q3: B
Honorcode: 1
Date: 3/30/00
Time: 10:20:17 PM
You can make statements about causation from the results of study two. This is because the groups are randomly taken out of the same population, Freshmen,
The Mean Square is actually the same as the population variance estimate, either between groups or within groups. When the Mean Square Between is equal to the Mean Square Within, the F ratio,(MSbetween / MSwithin) is 1. This shows that the null hypothesis is true and the between group estimat and the within group estimate are both estimates of the same population variance.
PCQ10_StudentID: 101835826
Nickname: lightning bolt
Q3: B
Honorcode: 1
Date: 3/30/00
Time: 10:20:57 PM
We can make statements about causation from the results of study two because this experiment is matching which is an overall better test. This kind of test is better because if the null hypothesis is true then the type one error is reduced. Also, it is better when the research hypothesis is true and the true differences among the groupd are large, making the power large. Study one is an example of a systematic selection which reduces the natural variation among samples to the extent that the variables on which the matching is being done are related to the variable being studied.
The term mean squares is another name for the variance because the variance is the mean of the squared deviations. The F ratio is the between-group estimate divided by the within-group estimate. Therefore, to make the Mean Square Between be equal to the Mean Square Within, the F ratio should be equal to one. In an analyis, a result of this in an experiment would be when the null hypothesis is true, the F ratio should be about 1, since the two population variances estimates are based on the same thing, the variance within each of the populations.
PCQ10_StudentID: 101175223
Nickname: rodimus
Q3: D
Honorcode: 1
Date: 3/30/00
Time: 10:44:24 PM
only the example of class and number of hours studied can be used to make statements about causation. if we had a comparison to gpa that was prior to the study, we could determine if there was a cause. the class study allows us to compare groups and determine if there is a cause for the number of hours studied
mean square is another term for variance, it is the average of the squared deviations. in order to get the mean square between equal to the mean square within, there must be about the same amount of deviation within the group as there is between the groups, with equal degrees of freedom as well. this would drastically reduce the F ratio to near 1.
PCQ10_StudentID: 101133209
Nickname: Elon
Q3: D
Honorcode: 1
Date: 3/30/00
Time: 10:47:25 PM
We can make causal statements about the second study because the Freshman in that study are "treated" by being assigned to different lengths of study-hall time per week. Therefore when their GPA's are measured at the end of the semester, we can make a causal statement about the length of study-hall time and GPA. Study one can not be a causal relationship because you can not state that your year in school causes the number of hours that you study a week. In addition, the 4 groups are not "treated" in any way, they are not given different assignments or classes or any thing to distinguish one from another in order to compare them.
The Mean Square is the variance as a measure of variabililty. It is a formula including the sum of squares divided by the degrees of freedom. The mean square between could be equal to the mean square within if the information found in each group is all the same and therefore the differences between would all be the same. If this happens, then the F ratio which is the mean squares between divided by the mean squares within will be 0. Therefore no causal statement can be made and no substantial relationship exists between the two groups.
PCQ10_StudentID: 100689400
Nickname: Chip Douglas
Q3: B
Honorcode: 1
Date: 3/30/00
Time: 10:53:57 PM
We are able to make statements about causation from the results of the second study. In this study, the different length study hall time per week (independent variable) is the cause or determinant of each group's semester GPA (dependant variable). This is the case because an opposite direction of causation is not possible. Semester GPA can't cause a person's length of study hall time because the experimenter contols the amount that each group spends in study hall per week.
A mean square is another name for the estimated population variance. It is equal to the sum of squared deviations from the mean divided by the degrees of freedom (SS/df). If you have samples from identical populations, in which there is little variation of the sample means within each population, you might get a Mean Square Between that is equal to the Mean Square Within. When the null hypothesis is true, the several populations being compared all have the same mean. In this situation, the within group estimate of the population variance and the between group estimate of the population variance are estimates of the same population variance. They should both be about the same. The resulting F ratio(MS Between/MS Within) would be 1.
PCQ10_StudentID: 101816052
Nickname: chou-chou
Q3: B
Honorcode: 1
Date: 3/30/00
Time: 10:56:58 PM
We can make statements about causation from the results of the second study because all of its participants are from the same group, freshmen, and it sets up different levels of the independant variable to be tested. The second study provides different lengths of study hall time. This makes it possible to make comments about causation from the lengths of study time to the GPA for that semester. The other study does not provide different levels of the independant variable.
A Mean Square is using variance as a measure of variability. There is a between group component which can be due to random error and treatment. There is also a within group component that can be due to random error. If there is no difference between the difference of the score and it's in-group mean and the difference of the score and it's grand mean, then the F ratio is equal to 1. This means that the treatment did not have an effect, the null hypothesis is true.
PCQ10_StudentID: 100747476
Nickname: GOAT
Q3: B
Honorcode: 1
Date: 3/30/00
Time: 11:07:53 PM
The experiment design is a controversial aspect of ANOVA testing. There are arguments on both sides of the issue as to create complete random selection to cerive variance analysis or to group samples to better reduce random variation. Study #2, since it is closer to complete randomness in selecting samples seems to be the better choice since it would most likely reduce variance.
Mean dquare is another way of saying variance. It is the sum of squared differences of scores. During ANOVA analysis, the mean square can be taken from either between groups or within groups. Within means that the variance is taken for a particular sample. It is calculated the same way as variance. (The squared sum of each score's deviation from the samople mean divided by degrees of freedom.) In ANOVA testing, each sample's variance is summerd and this gives the mean squares within. The mean square between is determined by summing the squared differences of each sample's mean less the grand mean, divided by degrees of freedom between. The mean of squares between might be equal to the mean of squares within if the scores are so similar that there is no visible difference. This would make it impossible to calculate an F score, since the F ration can not be 0.
PCQ10_StudentID: 100833872
Nickname: hairball
Q3: B
Honorcode: 1
Date: 3/30/00
Time: 11:09:43 PM
We can make statements about causation concerning study two. Within group or between group estimates can be caused by variations within populations or variations between populatons or both.
Mean square is another name for variance because the variance is the mean of the squared deviations. If the ratings in all the groups are similiar and the means of all the groups are similiar the mean square between has a better chance of being equal to the mean square within. The F ratio is the mean square between divided by the mean square within. If the mean square between is equal to the mean square within the ratio will be 1.
PCQ10_StudentID: 101566449
Nickname: Peaches
Q3: D
Honorcode: 1
Date: 3/30/00
Time: 11:13:28 PM
We can make statements about causation from the results of the second experiment because the independant variable can be controlled and manipulated and the GPA is what was affected by it and the differences or lack of differences in these means can be measured by the Anova tests.
The mean square is the variance. It is the mean of the squared deviations. If the mean square between was equal to the mean square within the F ratio would be equal to 1.
PCQ10_StudentID: 100951026
Nickname: shera
Q3: B
Honorcode: 1
Date: 3/30/00
Time: 11:18:33 PM
The first study uses freshman, sophomores, juniors and seniors and asks them how many hours a week they study. This is a simple analysis of variance study. With each group being totally independent from the others, if there is a huge difference in the variances between them and not a big variance within them, than the f ration would be far above 1 and we can conclude that there is a huge difference in the four groups' study times and a reason for the difference would be obviously the grade they are in. In the second group, the participants are all freshman and they are randomly selected and put in different length study halls. This study is not effective in saying what causes a high GPA, because who is to say the students actually study in their study hall time. Smart students may study in the library, so it wouldn't matter how long their study halls were. Some students who do badly in school and have a low GPA would do badly in a short or long study hall period because they don't study anyway. The point is that causality cannot be commented on in this study.
A mean square is a measure of variability. It is literally the Variance. MS=sum(X-M)^2 / n-1 or MS= SS/df. If there is the same amount of variance between the groups as there is within the groups, then the MSb and the MSw would be equal. If they were equal, then the Fobtained or F ratio would = 1 (Fobt=MSb/MSw). In other words, if the null hypothesis is true, then the the between group MS and the within group MS would be approximately equal and the Fobtained would = 1 and and you would consequently fail to reject Ho. The treatment would not have had an effect if the between group MS and the within group MS were equal.
PCQ10_StudentID: 101974684
Nickname: buttercup
Q3: B
Honorcode: 1
Date: 3/30/00
Time: 11:23:16 PM
Statements of causation can be made from the results of Study 2. Study 2 provides a random sample of freshman that are divided into 4 groups and assigned different lengths of study-hall time per week. The test is run in order to find out whether the amount of time spent studying can cause a certain semester GPA. The GPA seems to be solid dependent variable that can be found objectively, while the hours studied are carefully controlled. Since there are four different groups of freshman being used in the test, the results will be more accurate and will show the differences that the independent (or predictor) variable of time spent studying causes. With Study 1, a correlation, rather than a causation, could be shown between the class (Freshman, Sophomore, Junior, Senior) and the amount of time spent studying. However, it is unlikely that a such a strong relationship, one that would show causation, would be found between these two variables.
A mean square is another name for the variance because the variance is the mean of the squared deviations. In order to make the Mean Square Between be equal to the Mean Square Within in an experiment, the ratio between the two would have to be approximately 1 to 1. For instance, the Mean Square Between= 10/2, while the Mean Square Within= 5/1. In this situation, both values would be equal to 5. When both the estimate of the Mean Square Between and Mean Square Within are about the same, the null hypothesis is found to be true. When calculating the F ratio in this circumstance, F=5/5, F would equal 1. When F is close to 1, the null hypothesis is true.
What's up with number 1? I really couldn't find a good way to support my conclusion that Study 2 would be the test that could have causation. Help. :) Have a great day!
PCQ10_StudentID: 101435626
Nickname: Vokamis
Q3: B
Honorcode: 1
Date: 3/30/00
Time: 11:28:33 PM
The second study uses between group comparison. It measures the GPA of a similar group of people (freshmen) that have been assigned to four different experimentation group. Therefore, it would be more able to provide statements about causation.
The Mean Square is the same as the variance. Equal MSbetween and MSwithin would result in an F ration equal to 1.
PCQ10_StudentID: 100986806
Nickname: suzyq
Q3: D
Honorcode: 1
Date: 3/30/00
Time: 11:29:42 PM
You can make statements about causation on the second study only. The reason is that there are no levels to the dependent variable. The number of hours studied per week should be designated in different amounts by the experimentor.
Mean square is another term for the variance. If the null hypothesis is true than MS Between and MS Within are made up of the same variances, therefore they would be equal. In addition, because F=MS Between/MS Within F would then equal one.
PCQ10_StudentID: 101852531
Nickname: Daisy Girl
Q3: B
Honorcode: 1
Date: 3/30/00
Time: 11:34:55 PM
We can make statements of causation from Study two because we are manipulating the assigned study hall time in attempt to produce some difference in the dependent variable; the semester grade point average. You can't make a statement of causation for study one because there is no manipulation of the independent variable - you are just finding out if the means of the different group's study times are different - there is nothing in the experiment that is causing a change.
The Mean Square is the same thing as the population variance. The Mean Square of the Between group would be equal to the mean square of the within group if the individual scores varied within the groups the same as they did among the means of the groups - this would result in an F ratio of zero.
PCQ10_StudentID: 100881989
Nickname: scrubski
Q3: B
Honorcode: 1
Date: 3/31/00
Time: 12:15:56 AM
We can make statements about causation for the second study. Stating that studying more increases GPA. You cannot say that amount of hours studied determines what year you are in (study 1).
The mean square is the sum of squares divided by the degrees of freedom. When the two are equal teh F-ratio is 1
I will be unable to attend class tomorrow because I am taking this PCQ in Syracuse. I'll see you on Monday!!
PCQ10_StudentID: 101229881
Nickname: Artee
Q3: D
Honorcode: 1
Date: 3/31/00
Time: 12:29:06 AM
For study #1, causation can be made. The (x) independent variable, year in college, could be causing Y, the number of hours studied per week. However, in study 2, the (x) length of study hall time per week could be causing the Y semester GPA or vice versa, or some other factor may be causing both. Therefore, in study 2 it is difficult to come to a conclusion about which variable is really affecting the other.
A mean square is another name for the variance. This is because the variance is the mean of the squared deviations. The mean square between may be equal to the mean square within when the null hypothesis is true, they are both estimates of the same population variance. Therefore, the F-ratio will be 1 to 1 so the ratio is 1.
PCQ10_StudentID: 101229881
Nickname: Artee
Q3: D
Honorcode: 1
Date: 3/31/00
Time: 12:29:45 AM
For study #1, causation can be made. The (x) independent variable, year in college, could be causing Y, the number of hours studied per week. However, in study 2, the (x) length of study hall time per week could be causing the Y semester GPA or vice versa, or some other factor may be causing both. Therefore, in study 2 it is difficult to come to a conclusion about which variable is really affecting the other.
A mean square is another name for the variance. This is because the variance is the mean of the squared deviations. The mean square between may be equal to the mean square within when the null hypothesis is true, they are both estimates of the same population variance. Therefore, the F-ratio will be 1 to 1 so the ratio is 1.
Thanks for the cookies!!
PCQ10_StudentID: 101176381
Nickname: scooter
Q3: B
Honorcode: 1
Date: 3/31/00
Time: 12:46:20 AM
The Freshmen group that is put in to study periods is the only group that we can make statements about causation from. This is because the experiement is controlled with the two factors of study hall time and resulting GPA. In the other study, is not as controlled. The independent variable, class, is something that describes the participant, not a variable like that of study hall time. We can prove through the experiment that number of hours in study hall affects the GPA of the freshmen groups, but we cannot necessarily say that class causes hours spent studying. There would be too much variability in this group. Also It may be a coincidence that class and study time even relate. However, I would bet that the more time spent in the study hall would determine ones GPA.
Mean Squares is another term for the variance and can either be the MSbetween or the MSwithin. The F ratio is MSbetween/ MSwithin. If the null hypothesis is true then the variability between groups would equal variability within groups. As a result, the F ratio would be equal or close to 0. You would be unable to prove what you have set out to prove, which is often that the two groups differ in results. The groups would be equal and the resulting ratio would be right aroun 0.
PCQ10_StudentID: 101125631
Nickname: monkeygirl
Q3: D
Honorcode: 1
Date: 3/31/00
Time: 12:48:05 AM
You can make statements of causation about the second group, because you are setting up the groups by dividing 1 class (1 population) into different groups. Then by controlling how long each one studies, you are allowing n variance of study time within each group. This set-up allows you to make statements of causation.
The mean square is another name for the variance, because the variance is the mean of the squared deviation. Having equal MSW and MSB would give you a f-ration of 0.
for number three i picked "need more info" because to figure out the SS between group you need to know the sum of n, the group means, and the grand mean. You are given none of this information, therfore you cannot figure out what the SS between groups is.
PCQ10_StudentID: 101921521
Nickname: sweets
Q3: B
Honorcode: 1
Date: 3/31/00
Time: 12:53:00 AM
Statments about causation can be made about the second study by stating that studying more increases GPA. You can't, however, say that the amount of hours studied is the determining factor of what year you are in, which is study 1.
A Mean Square is the SS divided by the DF, the variance estimate. MS is, as usual, the same thing as S^2. However, in an analysis of variance table, the variance is almost always referred to as MS. The reason the mean square between and the mean square within are equal is due to the fact that there is no difference between the score and mean for the group and the score and mean for the total group. This affects the F ratio by making it equal to zero, and making the null hypothesis being true.
PCQ10_StudentID: 101142423
Nickname: mighty mouse
Q3: D
Honorcode: 1
Date: 3/31/00
Time: 12:56:25 AM
You can make statements about causation with the group that was divided into the four groups of freshmen. You can look at the GPAs and tell if they are higher or lower when they are compared, due to the length of the study hall.
A mean square represents the variance. The mean square between could be the same as the mean square when the null hypothesis is true. That means both estimates are based on the same thing. The F ratio should come out to be around 1.
PCQ10_StudentID: 101991256
Nickname: Figure8
Q3: D
Honorcode: 1
Date: 3/31/00
Time: 1:04:17 AM
Statements about causation can be made about the second study. The different length of study hall time can be associated with the semester GPA. Causes could be X to Y, Y to X and/or an alternate variable could have an impact. An important aspect of the second study that makes causation significant is that the independent variable includes populations/samples of the same level. In contrast, the independent variable in the first study includes populations/samples that have inconsistant or varied levels.
Causes could include samples which have the same mean-variance-shape. It would affect the F-ratio. The null hypothesis would be considered true which concludes that the experiment is invalid or had no effect.
PCQ10_StudentID: 101099628
Nickname: smurf
Q3: B
Honorcode: 1
Date: 3/31/00
Time: 1:06:27 AM
Statements about causation can be made for the results of the second ANOVA test with the four (4) groups of freshman assigned to study hall. This second study has the change in GPA as the dependent variable so we can observe whether the it is affected by the independent variable which is the different lengths of study hall time. Here the amount of study time manipulated. Unlike the first test where the amount of study time is not controlled.
A mean square is the sum of squares divided by the degrees of freedom. If the means and variances are equal among the populations then the mean square between and the mean square within will be equal or about the same. If this is so then the F ratio will be about 1.
=0}
PCQ10_StudentID: 101278100
Nickname: skibum
Q3: D
Honorcode: 1
Date: 3/31/00
Time: 1:20:07 AM
i would want to say that the one that you can make a causation from would be the one where the independant variable being study hall length becasue the it would allow causation on GPA. whereas with the other study the years that someone is in school may not have a difference on the amount of time that they study, and therefore would not provide a reason for causation.
the mean square is the population variance in the group or between the groups. In an ANOVA experiment something that may cause the mean square between and then mean square within to be the same might be if you were to be doing a study on the amount of rainfall within several areas and then by chance the means of the samples happened to be the same and then the amounts withing the populaitons were the same. What that would do the the F ratio would make it one. which would show that the populaitons were equal
PCQ10_StudentID: 100986806
Nickname: suzyq
Q3: B
Honorcode: 1
Date: 3/31/00
Time: 1:25:07 AM
answered earlier
answered earlier
I already sent in my pcq earlier tonight but I just realized that there was a way to figure out SS Between from the information given. I entered in the correct answer above!
PCQ10_StudentID: 100896105
Nickname: honeydo
Q3: D
Honorcode: 1
Date: 3/31/00
Time: 1:25:40 AM
You can make statements about the probably causation about the Freshman student GPA's being related to their study hours. The reason you can do this is because you are manipulating the amount of time studying for the groups of freshman. If there is a correlation, you can say that it was probably due to the studying but you can not say that this is not the causation is not guaranteed because other uncontrolable factors may have contributed to the outcome.
the mean squared is the SS/df. this is calculated for both within groups and between groups. The MSbg and MSwg can be equal if the same number of subjects were used making the dfs equal and then the SSbg and SSwg would have to be equal as well. If the MSbg and the MSwg were equal, the F obtained would be 1 because Fobt=MS bg/MSwg.
i do not fully understand the process and procedures for ANOVA. I think that we need to go over this more in class!
PCQ10_StudentID: 101105091
Nickname: sly
Q3: D
Honorcode: 1
Date: 3/31/00
Time: 1:36:59 AM
Study two can have statements made about causation because the different groups are from the same sample and the study controls the circumstances (number of hours studied). Therefore these results will be accurate and dependable.
-mean squares-the sample mean used to estimate the population variance -The MSbetween and the MSwithin might be the same if they are made up of the same variance sources. -if they were both the same then the F ratio would be 1
PCQ10_StudentID: 101945293
Nickname: rynoshaft
Q3: B
Honorcode: 1
Date: 3/31/00
Time: 1:44:07 AM
The freshman group can be used to make statements of causation because the group has little between group variance (Freshman most likely share similar classes and thus have similar demands placed on them) Meanwhile the other groups cannot be predicted because different classes may demand different amounts of study time.
Mean square is the square root of the sum of squares. When BG variance is approximately the same as within group variance it means that the independent variable does not account for the variance in the dependent variable. This means that the F ratio is close to one, and thus the null hypothesis is accepted
PCQ10_StudentID: 101244082
Nickname: applebum
Q3: D
Honorcode: 1
Date: 3/31/00
Time: 2:06:27 AM
We can make statements about causation only in situations where correlations can also be measured, so in the results of the Study 1, is the only time we could make causal statements. We cannot do it also for Study 2 because there are multiple groups in which various independent variables are assigned with the dependent variables.
The MEAN SQUARE is the estimate of population variance with-in groups and between groups. If the results of an experiment make the MS-BN equal to the MS-WN, that means that the population has little variation in the scores within it and therefore the means of samples from that population will be more likely similar. And this would affect the F ratio by making it equal 1.
On question 3, I do not understand what it means that effect size (R2) is .60 So far in all our situations and test statistics, the effect size has been determined by solving for Cohen's d and I originally thought that (R2) was the value of proportionate reduction error?. I am going to put D as the answer for question 3, but I do not understand it.
PCQ10_StudentID: 101137818
Nickname: giggle box
Q3: D
Honorcode: 1
Date: 3/31/00
Time: 2:18:32 AM
We are unable to make a statement of causation in the study hall time period and GPA experiment because there are too many other factors that could affect the students' GPA's such as the amount of time they spend at home studying. In the first experiment with the class and number of hours studied, we are able to make assumptions about causation because there are no underlying variables that would affect the outcome.
A mean square is another term used for variance because the variance is the mean of the squared deviations. The F ratio would be 1 if the mean square between and mean square within were equal.