Login to use UniMult now!

Login or Create a new Account
 
     
Forgot Login?   Sign up  

Walk-Through

Once you download and start UniMult, you will see a window titled, “Welcome to UniMult.” After agreeing to the conditions of use, you will then have the option of creating a new project or opening one of your recent projects, if any (as seen in the window below).



 

 

 

 

 

 

 

 

 

 

 

 

 

Next is the Data Screen window.On the right-hand side of this screen is a window titled “Tips for Data Entry.”Follow the 4 steps to enter your data (in the example below, the user has already entered in data). The last step on this page is to click “Next (Analyses),” which is located at the bottom right-hand corner of the window.Then you will get a pop-up that says that your data is saved with the .uMdat suffix.

Clicking "Next (Analyses)" will take you to the Analysis Screen window.It is made up of 3 tabs, Y(Dependent/Outcome) Variable(s), X(Independent/Predictor) Variable(s), and Variable Summary/Options, and another pop-up window titled “Tips for Data Analysis.”You are looking at the 3rd tab, which shows both the Y and X variables.Follow the 4 simple steps found on Tips for Data Analysisto analyze your data. (STEP 1: select your Y-Variables, STEP 2: select your X-Variables, STEP 3: select options, if any, and STEP 4: click "Next")

And that's all there is to simple analyses!

The program is sophisticated enough to know what kind of analysis is appropriate, does that analysis, and then labels the output correctly. For example, entering 2 Ys gives a multivariate analysis but entering 1 Y gives a univariate analysis. Options also appear on this screen when they are appropriate for the variables you selected.

When you have your data entered, play with the different possible analyses to learn more about the program—you cannot hurt it and will learn about your data too! If you are not sure about what the statistics mean, consult someone who knows or post it on the UniMult discussion board at unimult.com.

Below are the sample results of a tests run with this data set.

=====================================================================

User entered comments: Testing whether Religious Attendance is a function of Religious Importance and Intrinsic Motivation

=====================================================================

User entered comments: Testing whether Well Being varies by Age

Table x.

Analysis of Variance of Well Being with AgeNom

Effect Size

Variables r C. I. F Ratio df1 / df2 p

________________________________________________________________________

AgeNom .17 .00, .28 1.91 3 / 187 .13

_______________________________________________________________________

Notes: N = 191. C.I.'s are the lower and upper bounds for approximate 95%

confidence intervals. With df1 > 1, effect size is multiple correlation / eta.

 

Post Hoc Protected F and Scheffe F Test p's

Group 1 2 3 4

11 1. -- 0.56 0.37 0.22

12 2. 0.32 -- 0.52 0.31

13 3. 0.12 0.27 -- 0.44

14 4. 0.04 0.08 0.18 --

df1 = 1; df2 =186

The p's are two tailed. For a one tailed p, divide the p by 2.

=====================================================================

User entered comments: Evaluating whether the two church Camps differed in Age

Table x.

Cross-tab of Camp with AgeNom

Effect Size

Variables r C. I. F Ratio df1 / df2 p

________________________________________________________________________

AgeNom .31 .13, .41 6.50 3 / 187 .0003

_______________________________________________________________________

For a t-test, divide the p of any dichotomy by 2 (t = Square root of F with sign of r).

Notes: N = 191. C.I.'s are the lower and upper bounds for approximate 95%

confidence intervals. With df1 > 1, effect size is multiple correlation / eta.

 

Table x. Ns and Percentages for AgeNom By Camp

Camp

AgeNom 1 0 % N Odds Ratio

---------------------------------------------------------------

11 10 83% 2 17% 100% 12 5.00

% CIs 51, 96 --

12 78 87% 12 13% 100% 90 6.50

% CIs 78, 92 8, 22

13 72 88% 10 12% 100% 82 7.20

% CIs 79, 93 7, 21

14 2 29% 5 71% 100% 7 0.40

% CIs -- --

_______________________________________________________________

Notes: Total N = 191. Key for columns: 1. BCamp, 0. ACamp. Odds ratios are computed as the percentage in the first category divided by the second category. CIs are the approximate 95% confidence intervals.