Page 1 of 2 12 LastLast
Results 1 to 25 of 33

Thread: Homework Help - Statistics

  1. #1
    Featured Member tRoUbLeMaKeR's Avatar
    Joined
    Jan 2004
    Posts
    1,044
    Thanks
    16
    Thanked 9 Times in 8 Posts
    My Mood
    Amused

    Default Homework Help - Statistics

    I know this is a shot in the dark, but I'm working on my take home final for a stats class and I'm ready to pull my hair out! So I thought I'd at least ask.... I'm doing a factorial anova on SPSS and have to talk about what the F ratio means in relation to the data. All I know is that the higher the F the more variance is accounted for...but I don't know what this means or how to interpret a F value.

    If you know anything about this stuff please respond here or PM me. I'll be up late working on this

    thanks....and if no one responds in a day or two feel free to delete this post all together so it doesn't take up space

    thanks
    [SIGPIC][/SIGPIC]

  2. #2
    High_Heel_Lover
    Guest

    Default Re: Homework Help - Statistics

    If I help you, you will fail. Good luck! I got a head ache just reading this thread!

  3. #3
    Veteran Member
    Joined
    Apr 2006
    Location
    houston
    Posts
    513
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Default Re: Homework Help - Statistics

    I took 2 statistic classes and have no idea what you're talking about. (been a long time maybe). Can't you just copy the phrases your textbook uses in similar problems?

  4. #4
    Featured Member tRoUbLeMaKeR's Avatar
    Joined
    Jan 2004
    Posts
    1,044
    Thanks
    16
    Thanked 9 Times in 8 Posts
    My Mood
    Amused

    Default Re: Homework Help - Statistics

    Quote Originally Posted by High_Heel_Lover View Post
    If I help you, you will fail. Good luck! I got a head ache just reading this thread!
    LOL! Thanks High Heel Lover - you at least made me smile!


    No I wish I could just copy my textbook....but we aren't using one for this class. Just have my notes
    [SIGPIC][/SIGPIC]

  5. #5
    Veteran Member
    Joined
    Apr 2006
    Location
    houston
    Posts
    513
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Default Re: Homework Help - Statistics

    Search engines also work.

    First google result for statistics f ratio
    http://www.acastat.com/Handbook/14.html


    The F-ratio is used to determine whether the variances in two independent samples are equal. If the F-ratio is not statistically significant, you may assume there is homogeneity of variance

    Compare the test statistic with the f critical value (Fcv) listed in the F distribution. If the f-ratio equals or exceeds the critical value, the null hypothesis (Ho) (there is no difference between the sample variances) is rejected. If there is a difference in the sample variances, the comparison of two independent means should involve the use of the Cochran and Cox method.

    Is that the answer?

    Apparently if it is a small variance and small f value the standard t-test can be used to compare the means of two independent groups. Otherwise a more specialized method must be used.

  6. #6
    Featured Member sunnie's Avatar
    Joined
    Jul 2006
    Location
    The strip club...duh!
    Posts
    868
    Thanks
    0
    Thanked 69 Times in 9 Posts

    Default Re: Homework Help - Statistics

    I am working on the same thing!!!

    The F-test...so the test statistic is a ratio comes from the mean square between divided by the mean square within.

    Edited to add: In ANOVA for regression, which is what I was talking about in the SLR post below, f-test ratio is mean square residual divided by mean square error.

    Your null hypothesis in simple linear regression is that Bo=0, i.e. the variables have no linear relationship. If the Null Hypothesis is rejected, i.e. the p-value is in the critical region, then that means that there is a linear relationship between variables.

    Is that what you are asking about?
    Last edited by sunnie; 12-09-2007 at 07:38 PM.
    [SIGPIC][/SIGPIC]

  7. #7
    Featured Member sunnie's Avatar
    Joined
    Jul 2006
    Location
    The strip club...duh!
    Posts
    868
    Thanks
    0
    Thanked 69 Times in 9 Posts

    Default Re: Homework Help - Statistics

    So basically you are trying to pin down some of the variance in Y through its relationship with X. So, let's say the height of children is Y, and age is X. Some of the height in children in Y is accounted for by their age, but some is accounted for by other factors, genetics, nutrition, etc. If the null hypothesis is true, then none of the variation in children's heights can be explained by age. If we reject the null, then some of the variation in children's heights can be explained age.

    How much?

    R squared. What ever the value of r squared you get is the percent of variation in Y that can be explained by its linear relationship with X if you reject the null hypothesis that there is no linear relationship, Bo=0. (that is supposed to look like beta naught equals zero).

    R squared is also called the coefficient of determination.
    [SIGPIC][/SIGPIC]

  8. #8
    Veteran Member beautiful.'s Avatar
    Joined
    Aug 2007
    Posts
    645
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Default Re: Homework Help - Statistics

    Anybody else feel dumber than dirt after reading this thread?

  9. #9
    Featured Member sunnie's Avatar
    Joined
    Jul 2006
    Location
    The strip club...duh!
    Posts
    868
    Thanks
    0
    Thanked 69 Times in 9 Posts

    Default Re: Homework Help - Statistics

    Desavirsire can vouch that I have called her crying many times over my stats class. It is so motherfucking hard, I failed a midterm on it too. The professors go fast and if you don't catch on, you do really bad. I am one of the dumb ones I guess, but after writing out that post, I realize I have learned something.

    The first test I had in that class felt like a stats based IQ test more than a stats test.

    Edited to add:

    I just realized you are asking about something else. If you are tsting hypothesis about differences in group means, then the F-Test is a comparison of between group variability to within group variability. So the null is that there is an equality of all group means Ho: mean1 = mean2 = mean3 etc.

    You want to be able to reject the null so that you can go ahead and perform multiple comparison tests, like pairwise t-tests or whatever. If you have a large f-statistic, it will give you a small p value. Then you can go ahead and reject the null and move onto your next tests.
    Last edited by sunnie; 12-09-2007 at 07:34 PM.
    [SIGPIC][/SIGPIC]

  10. #10
    Featured Member tRoUbLeMaKeR's Avatar
    Joined
    Jan 2004
    Posts
    1,044
    Thanks
    16
    Thanked 9 Times in 8 Posts
    My Mood
    Amused

    Default Re: Homework Help - Statistics

    Ok so bare with me here. I obviously didn't learn much this semester.
    I'm doing this all on SPSS and it just gives me a summary table with the F value and the significance value. We didn't specify a specifc hypothesis.

    Maybe if I explain more about the specific problem it'll make more sense. The problem has 2 independent variables each with 2 levels, gender (male and female) and wine consumption (no wine or drinking wine). The dependent variable is food consumption. In the SPSS summary table it states the F value of wine consumption is 50.533 with a sig. value of .000. genders F value is .001 and has a sig. value of .981. The F value of the two combined is 5.08 with a sig. of .03.

    Based on this info I would just look at the significance values and determine obviously that wine consumption alone and wine consumption with gender is significant, whereas gender by itself is not. My homework asks me to interpret the F values in relation to the study like females drink more wine than men. I'm confused because I can gather this from other information besides the F value so I'm thinking I'm totally missing something!

    Argh! I could scream! And here I didn't go into brain surgery because I thought human services would be easier! ha!
    [SIGPIC][/SIGPIC]

  11. #11
    Veteran Member
    Joined
    Apr 2006
    Location
    houston
    Posts
    513
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Default Re: Homework Help - Statistics

    Maybe the F is measuring the variance of the groups? That is not the average or mean but how much men vary from other men that is, a wider curve compared to a steeper curve of distribution. Wine consumption has a wide variance aka more standard devitiations aka wider curve. It's been a long time since i took statistics.

  12. #12
    Banned ArmySGT.'s Avatar
    Joined
    May 2005
    Location
    SW Counter Troll HQ
    Posts
    5,582
    Thanks
    1,589
    Thanked 1,674 Times in 1,043 Posts
    Blog Entries
    13
    My Mood
    Amused

    Default Re: Homework Help - Statistics

    *surrounds self with ignorance*

  13. #13
    Featured Member sunnie's Avatar
    Joined
    Jul 2006
    Location
    The strip club...duh!
    Posts
    868
    Thanks
    0
    Thanked 69 Times in 9 Posts

    Default Re: Homework Help - Statistics

    Sorry, my info was useless... Aftering reading your second post, I realized we are doing different models. We haven't done factorial ANOVA, we learned it with generalized linear models. But I am soooooo happy to know there is a whole different family of models I get to learn later...oh the joys of parametric tests.

    I found this page that might help you....http://www.wadsworth.com/psychology_..._anova_01.html

    It was really good, I am glad I found it before finals, but they don't have a workshop on log-linear regression, which I could really use.
    [SIGPIC][/SIGPIC]

  14. #14
    Featured Member saphire123456's Avatar
    Joined
    Mar 2007
    Location
    Philly/AC/Vegas...
    Posts
    1,507
    Thanks
    16
    Thanked 129 Times in 80 Posts

    Default Re: Homework Help - Statistics

    ....thank you god, that i have passed statitistics without really understanding what all that crap means, and thank you again that i never have to willingly look at it again...
    thats all

  15. #15
    God/dess Bella21's Avatar
    Joined
    Sep 2003
    Location
    SoCal
    Posts
    7,772
    Thanks
    2
    Thanked 40 Times in 35 Posts

    Default Re: Homework Help - Statistics

    I had to learn that program for my major, too. Gah.

    For a factoral anova, it's used to assess differences between means, but in situations where you are interested in the effect of more than one independent variable. It allows you to assess the effect of individual variables (main effects) and the combined effect of more than one independent variable (interaction effects). The underlying logic is the same as in 1-way anova, but with the added interaction effect.

    For example, here is a results write-up for some mini project that I did a long time ago:

    Number of errors differed based on type of environment, F (1,36) = 141.92 p < .001. Mice housed in an enriched environment made fewer errors (M = 10.30, SD = 5.08 ) than mice housed in a standard environment (M = 18.00, SD = 4.29). No main effect was found for feeding schedule, F < 1, n.s. Mice with unlimited feeding made a similar number of errors (M = 14.00, SD = 8.33) as did mice on a once-a-day feeding schedule (M = 14.30, SD = 2.55). However, the main effect for environment was qualified by an environment by feeding schedule interaction, F (1,36)= 164.90, p < .001.
    If you think school is hard, try being stupid.

  16. #16
    God/dess
    Joined
    Dec 2004
    Location
    London, UK
    Posts
    2,218
    Thanks
    5
    Thanked 209 Times in 142 Posts

    Default Re: Homework Help - Statistics

    Try downloading a copy of Mintab from http://www.minitab.com/ You get a free 30 day trial.

    It'll crunch the stats for you and the help files will tell you something about the significance of the results.

    Sadly I'm too rusty on stats to be of any help without digging out my text books.

    Phil.

  17. #17
    Featured Member tRoUbLeMaKeR's Avatar
    Joined
    Jan 2004
    Posts
    1,044
    Thanks
    16
    Thanked 9 Times in 8 Posts
    My Mood
    Amused

    Default Re: Homework Help - Statistics

    First, let me say a big THANK YOU to everyone for their help and support! I've been working on this all day and am still getting through everything! It's due tomorrow so I'm staying up till I'm done.

    Bella21 - thanks so much for your input. I believe that's what I was looking for. You totally helped to guide me to where I needed to be because I was totally thinking I had to do something else! LOL!

    I do have another question though, hopefully Bella21 is on tonight

    I did an omnibus anova and then conducted two focused contrasts on spss. I need to estimate the effect size. as far as I know I can't do this on spss. I thought I need to do this with the eta formula but it comes out the same for each contrast and I don't think it's supposed to any advice?
    [SIGPIC][/SIGPIC]

  18. #18
    Yekhefah
    Guest

    Default Re: Homework Help - Statistics

    Quote Originally Posted by beautiful. View Post
    Anybody else feel dumber than dirt after reading this thread?

  19. #19
    Featured Member sunnie's Avatar
    Joined
    Jul 2006
    Location
    The strip club...duh!
    Posts
    868
    Thanks
    0
    Thanked 69 Times in 9 Posts

    Default Re: Homework Help - Statistics

    T-maker, good to hear you've made positive progress.

    Bella, I am totally asking you for help in the future!

    I have a quiz tomorrow on ANOVA for MLR and ANCOVA....blech!
    [SIGPIC][/SIGPIC]

  20. #20
    Featured Member tRoUbLeMaKeR's Avatar
    Joined
    Jan 2004
    Posts
    1,044
    Thanks
    16
    Thanked 9 Times in 8 Posts
    My Mood
    Amused

    Default Re: Homework Help - Statistics

    Good Luck Sunnie! I've got my take home due tomorrow and the in class portion tomorrow too! I'll be thinking of you! Let me know how it goes!
    [SIGPIC][/SIGPIC]

  21. #21
    Featured Member sunnie's Avatar
    Joined
    Jul 2006
    Location
    The strip club...duh!
    Posts
    868
    Thanks
    0
    Thanked 69 Times in 9 Posts

    Default Re: Homework Help - Statistics

    Good luck to you too, thanks!
    [SIGPIC][/SIGPIC]

  22. #22
    Veteran Member I_luv_dancers!'s Avatar
    Joined
    Nov 2006
    Location
    Somewhere Else
    Posts
    451
    Thanks
    0
    Thanked 1 Time in 1 Post

    Default Re: Homework Help - Statistics

    Good luck.

    Stats is a bitch. I got a D when I took it - but part of that I attribute to the professor, for whom English was a (poor) second language. If ever there was a class where information needs to be communicated clearly and concisely, it's Stats. Stats with a very heavy Korean accent ups the degree of difficulty exponentially.....
    Quote Originally Posted by Katrine View Post
    ....And I_Luv_Dancers.....can I say that your post brought a tear to my mildly buzzed eye? Right on...you are so awesome, thank you for the post. I really needed to hear it, esp. after what's happening with someone close to me. Big hugs and tugs to you!

    Quote Originally Posted by asianlady View Post
    ........I have had pretty good sex with hot guys and surprisingly great sex from over weight old farts who made me very vocal which I am not usually

  23. #23
    God/dess Bella21's Avatar
    Joined
    Sep 2003
    Location
    SoCal
    Posts
    7,772
    Thanks
    2
    Thanked 40 Times in 35 Posts

    Default Re: Homework Help - Statistics

    Quote Originally Posted by tRoUbLeMaKeR View Post
    I did an omnibus anova and then conducted two focused contrasts on spss. I need to estimate the effect size. as far as I know I can't do this on spss. I thought I need to do this with the eta formula but it comes out the same for each contrast and I don't think it's supposed to any advice?
    Gah, I wish I had more time for this thread right now, but I'm not at home and I won't be there 'till late (and my state of mine will probably not be suitable for stats)... I hope this is what you need!

    There are various ways of computing effect size depending on what type of hypothesis test you have run. If you mean a 1-way anova (and I sure hope you do, because I don't remember anything about "omnibus"... I actually had to google that term), then calculate r2 (squared). There are other measures of effect size for anova, but this one is appealing as it ranges from 0 to 1, and tells you the percentage of variability in your dependent variable that is related to your independent variable.

    A results example:

    F (2, 21) = 30.22, p < .0001, r2 = .74
    If you think school is hard, try being stupid.

  24. #24
    God/dess Bella21's Avatar
    Joined
    Sep 2003
    Location
    SoCal
    Posts
    7,772
    Thanks
    2
    Thanked 40 Times in 35 Posts

    Default Re: Homework Help - Statistics

    Ah yes... and I can't really type out the r-squared formula on this forum with the division line and all... I'm assuming you have it somewhere.

    It's r-squared = SSbetween over SStotal
    If you think school is hard, try being stupid.

  25. #25
    God/dess Bella21's Avatar
    Joined
    Sep 2003
    Location
    SoCal
    Posts
    7,772
    Thanks
    2
    Thanked 40 Times in 35 Posts

    Default Re: Homework Help - Statistics

    PS: I'm so fucking glad that I will never take another stats class in my life. BEGINNING stats was okay. Everything after that was an ugly wall of doom.
    If you think school is hard, try being stupid.

Page 1 of 2 12 LastLast

Similar Threads

  1. Homework Help
    By Autumn Lily in forum Stripping (was Stripping General)
    Replies: 7
    Last Post: 03-16-2011, 10:30 PM
  2. Homework Sucks
    By malloryknox in forum The Lounge
    Replies: 17
    Last Post: 09-11-2008, 12:38 PM
  3. Can you help me? Homework
    By High_Heel_Lover in forum The Lounge
    Replies: 18
    Last Post: 11-26-2007, 08:17 AM
  4. doing your homework
    By mermaidnz in forum Stripping (was Stripping General)
    Replies: 13
    Last Post: 01-07-2005, 09:15 AM
  5. Homework help
    By LayaLeighton in forum The Lounge
    Replies: 15
    Last Post: 01-02-2004, 11:17 AM

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •