Statistics Help- Hypothesis Testing

earthfell · 05-07-2008, 09:00 PM

I need help!!!

Can someone explain hypothesis testing to me, or link a website they think explains it clearly?

I HAVE to pass this class or my transfer admission will be revoked, and everything for the rest of the month is based on hypothesis testing!!!

Vehn · 05-07-2008, 09:04 PM

Statistical hypothesis testing - Wikipedia, the free encyclopedia

JeydaX · 05-07-2008, 09:34 PM

The hardest part about hypothesis testing is by far the "statistical explanations" you have to use at the very end of the test... at least that's what the hardest part was for me.

What kind of hypothesis testing are you doing? There are several different ones. Are they z-score tests, t-tests, proportion tests? I could probably help at least a little bit (I am finishing up my third stats course with my final coming this Tuesday at the UofMinnesota.)

Rune · 05-07-2008, 09:39 PM

Buy the textbook. Open it. Work some problems. If you don't understand the problems in the current chapter, go back until you do. You're never going to understand it otherwise.

voodoochile78 · 05-07-2008, 09:59 PM

If you want to succeed in statistics, you had better get comfortable with the idea that a sample mean(different than the population mean) has its own distribution (Gaussian for reasonably large samples).

Once you get that, then it's as simple as integrating a Gaussian distribution from one point to another.

If you're asking about hypothesis testing, I'm guessing that you don't have that instinctual understanding of the fact that sample means have their own distribution.

Schatze · 05-07-2008, 10:22 PM

1.) Dichotomize populations into the subjects in the test and and normal group (teens in this study who wear hoodies and their liking of Linkin Park), and individuals who are not like those in the study.

State research and null hypothesis. Depending upon the method you're using, this may be as simple as mu1 != (use dashed equals sign) mu2 (the fancy U symbol that means population).

2.) Determine the characteristics of your distribution;
What distribution they'll fit in (T, F, Z, Chi square). Also the degrees of freedom if that is applicable. Some profs will want you to put other bits of information here

3.) Determine the cutoff score for your particular data set using your degrees of freedom, number of participants, alpha (.05, .01). Whether the test is one tailed or two tailed, if that particular distribution and type allows for it. Something like T critical one tailed = xxx.xx(df1), p. < .05, xxx.xx is looked up on a table

4.) Do your calculations here (or in space provided), to determine if your sample's score exceeds the critical value for the cutoff as found on a handy table.

5.) Decide whether to reject the null hypothesis; i.e. teenagers in hoodies do listen to Linkin Park at a greater level than the population at large

If you can't manage to memorize what you do where in a 5 step hypothesis test, you're going to be in major shit remembering the formulas. And if you need someone to hold your hand through explaining how each of the 5 steps works for T tests of various sorts, F tests, Z test, ANOVA, Chi-Square test, linear regression, etc., you're boned, because I went to classes and got As, so I have little sympathy for someone who probably skipped :P

I'm honestly confused though, you mention transferring so I assume you mean university, but a month on the 5 steps of hypothesis testing? Wut? We spent 15 minutes on it and then applied it for all the various parametric and a couple non-parametric statistical methods. THat was 2101 stats. Advanced lab stats was parametric and non-parametric stats with a heavy focus on getting to know and love SPSS.

P.S. It might be helpful to say WHAT stats course you're in. I've done stats for psychology which is very, very different from stats for engineering, or even business stats. Also that wiki page sort of sucks as it uses computational formulas which you won't be using in modern stats class. Computational formulas are sort of useless in the modern teaching of stats.

OneofOne · 05-09-2008, 01:39 AM

Do what I did - make friends with someone who "gets it" and leech. Stats is by far the worst class I ever took - I hated it with every fiber of my being. My professor wrote the book that we used which at least made it a bit easier. Still, much hate. GL!

*fake edit - Schatze has posted the standard layout you can use, and you can fill a lot of space by including all the various graphs, annotated

earthfell · 05-21-2008, 12:50 AM

Quote:

Originally Posted by Schatze

1.) Dichotomize populations into the subjects in the test and and normal group (teens in this study who wear hoodies and their liking of Linkin Park), and individuals who are not like those in the study.

State research and null hypothesis. Depending upon the method you're using, this may be as simple as mu1 != (use dashed equals sign) mu2 (the fancy U symbol that means population).

2.) Determine the characteristics of your distribution;
What distribution they'll fit in (T, F, Z, Chi square). Also the degrees of freedom if that is applicable. Some profs will want you to put other bits of information here

3.) Determine the cutoff score for your particular data set using your degrees of freedom, number of participants, alpha (.05, .01). Whether the test is one tailed or two tailed, if that particular distribution and type allows for it. Something like T critical one tailed = xxx.xx(df1), p. < .05, xxx.xx is looked up on a table

4.) Do your calculations here (or in space provided), to determine if your sample's score exceeds the critical value for the cutoff as found on a handy table.

5.) Decide whether to reject the null hypothesis; i.e. teenagers in hoodies do listen to Linkin Park at a greater level than the population at large

If you can't manage to memorize what you do where in a 5 step hypothesis test, you're going to be in major shit remembering the formulas. And if you need someone to hold your hand through explaining how each of the 5 steps works for T tests of various sorts, F tests, Z test, ANOVA, Chi-Square test, linear regression, etc., you're boned, because I went to classes and got As, so I have little sympathy for someone who probably skipped :P

I'm honestly confused though, you mention transferring so I assume you mean university, but a month on the 5 steps of hypothesis testing? Wut? We spent 15 minutes on it and then applied it for all the various parametric and a couple non-parametric statistical methods. THat was 2101 stats. Advanced lab stats was parametric and non-parametric stats with a heavy focus on getting to know and love SPSS.

P.S. It might be helpful to say WHAT stats course you're in. I've done stats for psychology which is very, very different from stats for engineering, or even business stats. Also that wiki page sort of sucks as it uses computational formulas which you won't be using in modern stats class. Computational formulas are sort of useless in the modern teaching of stats.

Ok, so after being thoroughly admonished in this thread I went back and now I can do hypothesis testing.

Except for one tricky little detail.

Step 5! I know how to reject or fail to reject, but the statement I have to make afterwards confuses me. I am not sure how to phrase it.

Say I am testing a claim that drug D is effective in over 80% of cases. My H0: p=0 and H1: p>.8
I do steps 1-4 and find that my Z score is not within the critical region and so I fail to reject. What would my conclusion be and why? What if the z score was in the critical region and I rejected the H0?

I am taking introductory statistics. It is an online class (through the junior college) so I have to self teach. Basically what happened is I found out I got into berkeley, got psyched and spent a week doing nothing but paperwork/scholarships, and fucked myself over with chapter 8 (hypothesis testing). All i needed was a little explaining from someone who understands it, which i dont get much of for this course.

Right now I am starting chapter 10 (9 was "inferences from two samples" and 10 is "correlation and regression"). I only have two weeks before finals so I dont have a lot of time, which is why i am asking for assistance here for the things i am fuzzy with.

Thank you for the help!

Millie · 05-21-2008, 06:25 AM

I haven't gotten to hypothesis testing yet. I think that's coming up in next week's class for me. But when it does...ouch. Not looking forward to it.

Schatze · 05-21-2008, 10:55 AM

Stats is the big EZ. If you aren't afraid of it, and bother to look at the logic behind it, it's not too bad. Especially if you're not doing a highly technical stats course like I did. The people who failed miserably, and a lot of people do fail stats, or do poorly, are the ones who don't bother to grasp the logic and the concepts but just look at it as writing out formula.

One tailed test, Z distribution. Pretend that looks like a normal distribution, the infamous bell curve.
_______. ^ .
______._____.
____. _______ .
_ . ___________.
/__|__|__|__|__|_\
-2SD-1SD M____^

Anything left of the arrow would be below critical threshold and you would fail to reject the null. Anything right of the arrow would be more extreme than expected depending upon your significance level. I've arbitrarily placed it near where two standard deviations above the mean would be because that cuts off 96% of the distribution, so close to .05 (or would in a normal distribution). Now, if you're doing a one tailed hypothesis test, you'd want your Z score to be to the right of the arrow. That is, that your Z score is more extreme than the critical value. If it is less extreme than the critical value, you will fail to reject the null. If it is more extreme, you reject the null hypothesis and conclude that the results are significant. The critical value corresponds to a Z score with arbitrary that is given in your tables for the cutoff for the particular significance level cutoff you're interested in. So if your results are more extreme than the critical cutoff, your reject the null and you've "proven" whatever it is you're trying to prove. If the results are less extreme, you cannot reject the null and the results are not significant.

In your step 5.) you may want to add a caveat by including a "significantly" in your conclusion. It depends on how anal the prof is. There's always a chance of error, so you can't say definitively one way or the other(p. < .05 is less than 5% chance of error, .01 less than 1% etc.); however, when speaking within the context of the test, you *did* prove it at your predetermined significance level.

If you fail to reject the null, then it means there was a statistically significant difference.

In my above example, "(test type) crticial was exceeded, therefore the null is rejected. Students who wear hoodies listen to more Linken Park then the population at large."

BTW, your p value is wrong I think. Since you're testing the null it would be p. < .2, that is if the p value is larger than .2 that means the claim is false, whereas if the claim is true the p value would be less than .2, cutting off 80% of the distribution. Remember you're testing the null hypothesis which is basically the opposite of the research hypothesis. But I'm not certain what type of stats course you're doing and how the questions are phrased so you may want to double check that you're doing it right by doing the practice problems and making sure your answers are correct. I assume you're using H to signify mu.

Correlation and linear regression: correlation is easy, you'll probably be doing the pearson product moment correlation and not a non-parametric like Spearman's rho. so you'll be deriving r.

linear regression is easy too as long as you keep your concepts straight like Y^ (should be above the Y, called Y hat). Depending how in depth you'll get it'll look something like Y^ = a + b(X1)(X2)(X3), for however many predictor variables you have (the X axis is the predictor variable(s), Y the criterion variable). a is the correlation constant which is the Y intercept, where the regression line crosses the Y axis, b is the regression coefficient which is the slope of the line. Basically that allows you to predict what your expected Y value (Y^) will be for any given X, or vice versa, which falls along the regression line.

If you get in depth, you may go onto the general linear prediction rule from which most parametric statistics are derived. This includes error, i.e. how much do the predictors actually predict the Y variable. For instance, like if you wanted to predict how well a person would do in his stats class the first exam, the 3rd exam, and the amount of homework turned in accounted for 56% of determining outcome of final mark (actual example), with 44% of how well they did being unaccounted for by those predictors.

edit: the stats courses I've done have been highly technical and theoretical, rather than "understanding stats". As in, I could do the statistical calculations or examine the statistical calculations of a published paper. I used A. Aron in my 2nd year stats, chapter one was descriptive, chapter two was explaining what inferential statistics are, 3 was T, 4 was T using n > 1, 5 Z, then each chapter to a different statistical method. Hypothesis testing was part of chapter 2. This was 2nd year stats. 3rd year was non-parametric and using SPSS mainly.

Question: what sort of stats course does hypothesis testing come near the end? I realize I was taking a course for stats in a particular branch of an academic field, but I can't imagine how you would only get around to hypothesis testing near the end rather than the beginning. Was your course more about descriptive statistics than inferential statistics?

earthfell · 06-02-2008, 10:49 AM

My final is tonight!

Thanks for all the help; I have a very strong understanding of hypothesis testing and will be walking into the room with lots of confidence!

The final is cumulative, with half of the points being old stuff (chp 1-8) and the other half being new stuff (chp9-12)

The new stuff.... chapter 9 was on the comparison of two populations. I can have a hypothesis test on two proportions, two independent means, matched pairs (two dependent means), or two variances. For the mean tests, standard deviation can be known or unknown. If they are known, I use the Z distribution, if they are unknown I use the T distribution. Std deviation used the F distribution.

Chapter 10 deals with regression and correlation.... too much for me to want to recap it, but you summed it up nicely above.

Chapter 11 I felt was one of the easiest chapters. I just test for independence, homogeneity, or goodness of fit... all of which have f-distr with chi-square.

Chapter 12 was ANOVA.... she said there would only be one question on it, and it is really easy anyway.

All of the new stuff was hypothesis testing. SO glad I get the concept now.

If you need any help let me know Millie....

05-07-2008, 09:00 PM	#1 (permalink)
earthfell Registered User Join Date: Mar 2007 Posts: 275 -4 Internets	Statistics Help- Hypothesis Testing I need help!!! Can someone explain hypothesis testing to me, or link a website they think explains it clearly? I HAVE to pass this class or my transfer admission will be revoked, and everything for the rest of the month is based on hypothesis testing!!!

05-07-2008, 10:22 PM	#6 (permalink)
Schatze Registered User Join Date: May 2002 Posts: 1,374 -4 Internets	1.) Dichotomize populations into the subjects in the test and and normal group (teens in this study who wear hoodies and their liking of Linkin Park), and individuals who are not like those in the study. State research and null hypothesis. Depending upon the method you're using, this may be as simple as mu1 != (use dashed equals sign) mu2 (the fancy U symbol that means population). 2.) Determine the characteristics of your distribution; What distribution they'll fit in (T, F, Z, Chi square). Also the degrees of freedom if that is applicable. Some profs will want you to put other bits of information here 3.) Determine the cutoff score for your particular data set using your degrees of freedom, number of participants, alpha (.05, .01). Whether the test is one tailed or two tailed, if that particular distribution and type allows for it. Something like T critical one tailed = xxx.xx(df1), p. < .05, xxx.xx is looked up on a table 4.) Do your calculations here (or in space provided), to determine if your sample's score exceeds the critical value for the cutoff as found on a handy table. 5.) Decide whether to reject the null hypothesis; i.e. teenagers in hoodies do listen to Linkin Park at a greater level than the population at large If you can't manage to memorize what you do where in a 5 step hypothesis test, you're going to be in major shit remembering the formulas. And if you need someone to hold your hand through explaining how each of the 5 steps works for T tests of various sorts, F tests, Z test, ANOVA, Chi-Square test, linear regression, etc., you're boned, because I went to classes and got As, so I have little sympathy for someone who probably skipped :P I'm honestly confused though, you mention transferring so I assume you mean university, but a month on the 5 steps of hypothesis testing? Wut? We spent 15 minutes on it and then applied it for all the various parametric and a couple non-parametric statistical methods. THat was 2101 stats. Advanced lab stats was parametric and non-parametric stats with a heavy focus on getting to know and love SPSS. P.S. It might be helpful to say WHAT stats course you're in. I've done stats for psychology which is very, very different from stats for engineering, or even business stats. Also that wiki page sort of sucks as it uses computational formulas which you won't be using in modern stats class. Computational formulas are sort of useless in the modern teaching of stats. Last edited by Schatze : 05-07-2008 at 10:51 PM.

05-21-2008, 10:55 AM	#10 (permalink)
Schatze Registered User Join Date: May 2002 Posts: 1,374 -4 Internets	Stats is the big EZ. If you aren't afraid of it, and bother to look at the logic behind it, it's not too bad. Especially if you're not doing a highly technical stats course like I did. The people who failed miserably, and a lot of people do fail stats, or do poorly, are the ones who don't bother to grasp the logic and the concepts but just look at it as writing out formula. One tailed test, Z distribution. Pretend that looks like a normal distribution, the infamous bell curve. _______. ^ . ______._____. ____. _______ . _ . ___________. /__\|__\|__\|__\|__\|_\ -2SD-1SD M____^ Anything left of the arrow would be below critical threshold and you would fail to reject the null. Anything right of the arrow would be more extreme than expected depending upon your significance level. I've arbitrarily placed it near where two standard deviations above the mean would be because that cuts off 96% of the distribution, so close to .05 (or would in a normal distribution). Now, if you're doing a one tailed hypothesis test, you'd want your Z score to be to the right of the arrow. That is, that your Z score is more extreme than the critical value. If it is less extreme than the critical value, you will fail to reject the null. If it is more extreme, you reject the null hypothesis and conclude that the results are significant. The critical value corresponds to a Z score with arbitrary that is given in your tables for the cutoff for the particular significance level cutoff you're interested in. So if your results are more extreme than the critical cutoff, your reject the null and you've "proven" whatever it is you're trying to prove. If the results are less extreme, you cannot reject the null and the results are not significant. In your step 5.) you may want to add a caveat by including a "significantly" in your conclusion. It depends on how anal the prof is. There's always a chance of error, so you can't say definitively one way or the other(p. < .05 is less than 5% chance of error, .01 less than 1% etc.); however, when speaking within the context of the test, you did prove it at your predetermined significance level. If you fail to reject the null, then it means there was a statistically significant difference. In my above example, "(test type) crticial was exceeded, therefore the null is rejected. Students who wear hoodies listen to more Linken Park then the population at large." BTW, your p value is wrong I think. Since you're testing the null it would be p. < .2, that is if the p value is larger than .2 that means the claim is false, whereas if the claim is true the p value would be less than .2, cutting off 80% of the distribution. Remember you're testing the null hypothesis which is basically the opposite of the research hypothesis. But I'm not certain what type of stats course you're doing and how the questions are phrased so you may want to double check that you're doing it right by doing the practice problems and making sure your answers are correct. I assume you're using H to signify mu. Correlation and linear regression: correlation is easy, you'll probably be doing the pearson product moment correlation and not a non-parametric like Spearman's rho. so you'll be deriving r. linear regression is easy too as long as you keep your concepts straight like Y^ (should be above the Y, called Y hat). Depending how in depth you'll get it'll look something like Y^ = a + b(X1)(X2)(X3), for however many predictor variables you have (the X axis is the predictor variable(s), Y the criterion variable). a is the correlation constant which is the Y intercept, where the regression line crosses the Y axis, b is the regression coefficient which is the slope of the line. Basically that allows you to predict what your expected Y value (Y^) will be for any given X, or vice versa, which falls along the regression line. If you get in depth, you may go onto the general linear prediction rule from which most parametric statistics are derived. This includes error, i.e. how much do the predictors actually predict the Y variable. For instance, like if you wanted to predict how well a person would do in his stats class the first exam, the 3rd exam, and the amount of homework turned in accounted for 56% of determining outcome of final mark (actual example), with 44% of how well they did being unaccounted for by those predictors. edit: the stats courses I've done have been highly technical and theoretical, rather than "understanding stats". As in, I could do the statistical calculations or examine the statistical calculations of a published paper. I used A. Aron in my 2nd year stats, chapter one was descriptive, chapter two was explaining what inferential statistics are, 3 was T, 4 was T using n > 1, 5 Z, then each chapter to a different statistical method. Hypothesis testing was part of chapter 2. This was 2nd year stats. 3rd year was non-parametric and using SPSS mainly. Question: what sort of stats course does hypothesis testing come near the end? I realize I was taking a course for stats in a particular branch of an academic field, but I can't imagine how you would only get around to hypothesis testing near the end rather than the beginning. Was your course more about descriptive statistics than inferential statistics? Last edited by Schatze : 05-21-2008 at 12:53 PM.

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search
Display Modes	Rate This Thread
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode	Rate This Thread:

05-07-2008, 09:04 PM	#2 (permalink)
Vehn Hurm... Join Date: Jun 2007 Posts: 1,449 +1 Internets	Statistical hypothesis testing - Wikipedia, the free encyclopedia

05-07-2008, 09:39 PM	#4 (permalink)
Rune Bonafied Misanthrope Join Date: Oct 2004 Location: ATX Posts: 901 -2 Internets	Buy the textbook. Open it. Work some problems. If you don't understand the problems in the current chapter, go back until you do. You're never going to understand it otherwise.

05-07-2008, 09:59 PM	#5 (permalink)
voodoochile78 Registered User Join Date: Apr 2006 Posts: 638 -20 Internets	If you want to succeed in statistics, you had better get comfortable with the idea that a sample mean(different than the population mean) has its own distribution (Gaussian for reasonably large samples). Once you get that, then it's as simple as integrating a Gaussian distribution from one point to another. If you're asking about hypothesis testing, I'm guessing that you don't have that instinctual understanding of the fact that sample means have their own distribution.

05-09-2008, 01:39 AM	#7 (permalink)
OneofOne Registered User Join Date: Nov 2006 Posts: 313	Do what I did - make friends with someone who "gets it" and leech. Stats is by far the worst class I ever took - I hated it with every fiber of my being. My professor wrote the book that we used which at least made it a bit easier. Still, much hate. GL! *fake edit - Schatze has posted the standard layout you can use, and you can fill a lot of space by including all the various graphs, annotated

05-21-2008, 06:25 AM	#9 (permalink)
Millie Feisty Join Date: Jan 2002 Posts: 5,715	I haven't gotten to hypothesis testing yet. I think that's coming up in next week's class for me. But when it does...ouch. Not looking forward to it.

06-02-2008, 10:49 AM	#11 (permalink)
earthfell Registered User Join Date: Mar 2007 Posts: 275 -4 Internets	My final is tonight! Thanks for all the help; I have a very strong understanding of hypothesis testing and will be walking into the room with lots of confidence! The final is cumulative, with half of the points being old stuff (chp 1-8) and the other half being new stuff (chp9-12) The new stuff.... chapter 9 was on the comparison of two populations. I can have a hypothesis test on two proportions, two independent means, matched pairs (two dependent means), or two variances. For the mean tests, standard deviation can be known or unknown. If they are known, I use the Z distribution, if they are unknown I use the T distribution. Std deviation used the F distribution. Chapter 10 deals with regression and correlation.... too much for me to want to recap it, but you summed it up nicely above. Chapter 11 I felt was one of the easiest chapters. I just test for independence, homogeneity, or goodness of fit... all of which have f-distr with chi-square. Chapter 12 was ANOVA.... she said there would only be one question on it, and it is really easy anyway. All of the new stuff was hypothesis testing. SO glad I get the concept now. If you need any help let me know Millie....