4 Recommendations

All Answers (46)

Deleted profile

two-way ANOVA Is not possible in any version of SPSS. Its upto single only.

1 Recommendation

5th May, 2012

Gudrun Smith

Technische Universität Braunschweig

Dear Mr Singh

Do you have an idea with which programme it is possible to do a two-way ANOVA?

Thank you for your help in advance.

5th May, 2012

Dana Sugu

Jadavpur University

http://www.wessa.net/rwasp_Two%20Factor%20ANOVA.wasp

Deleted profile

Mr. Smith, I will let u know very shortly.

5th May, 2012

Shashi Ajit Chiplonkar

Jehangir Hospital

There is a typo mistake in my question. I wish to test the variation due to region and varieties. is there any other stats software giving two-way anova?

1 Recommendation

5th May, 2012

Diogo Pinho

BIOCANT

Hi

It is possible using the SPSS for performing a two-way ANOVA. It is difficult explain here. If you want I can take a look at the data because you have to decide if you are going to do two-way ANOVA of independent samples or repeated or mixed factors (an independent factor and repeated). For each ANOVA is necessary verify certain premises required for ANOVA (depending on the type of samples).

Regards,

Diogo Pinho

1 Recommendation

Deleted profile

Mr. Diogo, Kindly write the path of two-way ANOVA on SPSS.

5th May, 2012

Diogo Pinho

BIOCANT

Hi Harpreet Singh,

Independent samples: analyse/general linear model/univariate > fill the fields; in options select: descriptive statistics, estimates of effect size, observed power and homogeneity tests. in save select: residual unstandardized. IF you reject the null hypothesis: analyse/general linear model/univariate; in post hoc select: bonferroni and tukey.

Repeated measures: analyse/general linear model/repeated measures > fill the fields; in options select: descriptive statistics, estimates of effect size, observed power/continue; in save select: residual unstandardized. IF you reject the null hypothesis: analyse/general linear model/repeated measures; In options move variables for "display means for"; select compare main effects and confidence interval adjutements; select bonferroni/continue/OK.

If you need anything else please do not hesitate to contact me.

Regards,

Diogo Pinho

2 Recommendations

5th May, 2012

Shashi Ajit Chiplonkar

Jehangir Hospital

Hello Mr Pinho

I have 5 regions and 10 varieties and one observations per cell for protein content. How I can fill this data in a general linear model fields? I wish to estimate between regions variance, between varieties variance and test them against the error variance.

5th May, 2012

Diogo Pinho

BIOCANT

If I understand correctly you need to build a matrix similar to the attached. Then you have to apply a two-way ANOVA for independent samples, since region and variety are independent samples right?

If you send me the data would be easier to explain to you because using the little information that you have provided becomes difficult assess about which method should be used.

Regards,

Diogo Pinho

• example.xlsx9.36 KB

1 Recommendation

5th May, 2012

Shashi Ajit Chiplonkar

Jehangir Hospital

Attaching data column in your excel sheet.

• example.xlsx9.30 KB

5th May, 2012

Anand A Khot

Bioseed Research India Pvt. Ltd.

@Diogo p and @All

It;s nice to read your discussion. Can you provide me NTSYS software and it's flow chart if you peoples are using it????????

5th May, 2012

Rani Yadav

Shivaji University, Kolhapur

Test Procedure in SPSS:

Click Analyze ==> General Linear Model ==> Univariate...

Select the dependent variable, and select the fixed factors.

Click the “Model…” button. In the Univariate:Model window, select the “Custom” option and then the pull-down option in the center for “Main effects”. Select “factor_A”

and “factor_B” and move them to be in the Model. Next, after changing the pull-down option in the center to “Interaction”, select both “factor_A” and “factor_B” and then move them

to be in the Model.

1 Recommendation

5th May, 2012

Elijah Oyoo-Okoth

Analynk Data Solutions

Hi Sashi

To analyze data in 2 way ANOVA is very simple in SPPS even without seeing your data - I assume you dont have advanced knoweldge of SPSS, so pardon me if I repeat what you already know, (I assume you have SPSS ver. 12 to 19)

Preliminary information (Step 1): Enter your data in SPSS data sheet with the first variable representing regions (e.g 1 = Pradesh, 2 = New York; 3 = Casablanka, 4 = Nairobi etc) and another variable 1 to n representing varieties (e.g. 1 = Simpson; 2 = Calmony, 3 = Malachite, 4 = Obama etc). These variable 1 and 2 will be your FACTORS (VERY IMPORTANT). NOTE = the data must be Numeric for any analysis to proceed!!

Step 2: Enter a third variable for protein levels (e.g. 1 = 20%, 2 = 25%, 3 = 40%, 4 = 60% etc). = This will be DEPENDANT VARIABLE

STEP 3: Click Analyze ==> General Linear Model===> click Univariate. Sashi once there you can see FACTORS AND DEPENDENT VARIABLE very easily. Click Protein levels into the DEPENDANT VARIABLE BOX; Then region and variety will be clicked into the FIXED FACTOR(S). Since you dont have any other factors likely to affect your results e.g. temperature variations then dont worry about the other boxes.

STEP 4: Click Model and change from Full factorial to custom (Very crucial step because that is where you can get your interactive ANOVA)

STEP 5: While keeping the Ctrl Key in the computer pressed down, click regions and varieties (Ctrl key only ensure that you select the 2 factors at one go), and check them into the Model box {Sasha if you see Built term(s), Type change from the default Interactions to All 2-Way, Click continue and Voila You Click OK and you have your 2-Way ANOVA outputs.

4 Recommendations

5th May, 2012

Shashi Ajit Chiplonkar

Jehangir Hospital

Thanks a lot Dr Elijah Oyoo-Okoth. I could get the two-way ANOVA as you illustrated. But by enetering region and varieties at one go, does not give main effects of region and variety. Instead if I entered both seperately and interactions to all 2-way, I got the ANVOA table for 2-way ANOVA. However I did not understand what is the Corrected model? and intercept?

Thanks once again.

Shashi

5th May, 2012

Elijah Oyoo-Okoth

Analynk Data Solutions

Dear Sashi

The discussion is quite interesting. Sometimes the principles behind statistical procedures and the algorithm involved are complex, that we may not really understand them at some levels. However, ANOVA work with means. Main Effects is just the effect of one factor since you have 10 regions, replicated 5 times (5 varieties - ANOVA will be calculated); Similarly, you have 5 varieries replicated 10 times (by 10 regions) (It is easy to calculate ANOVA since ,means and variance can be calculated). After running your data, I could not get 2-WAY ANOVA; I guess what you had was Main effect ANOVA for region and Variety. For the Interaction (Interactions can be 2-Way, 3-Way, 4-Way, 5-WAy - So simply go to 2-way ANOVA). I doubt if you got F and p-value for Interactions. There is a shortcoming for 2 way ANOVA based on your data (Interacting data sets lack replication!!).

Two way ANOVA need the variation for example Region 1 has only 1 (instead of 2 or more) variety 1 (1 interacting data point). If you were to calculate the mean for the regions per variety, then you will end up with all your data as they are currently constituted!!!. E.g. calculate mean (and variance) for region 1 var. 1 (You cannot because, it is only 1 interacting observation). So please if you want to run and get F value and p values of Two-Way ANOVA, then add the third column for replicates so as to create a variation in your (interaction) data for Factor 1 and Factor 2. For example 1.82 was expected to be mean protein for region 1 and variety 1 (It will be a mean with a standard deviation - loosely we can say it has created a variance). Check below. If you can replicate your data (its unfornuate that you have to go back to the experiment, then you can fill colums with the ?????, which will generate raw data to calculate the means, then runnnig 2-WAy ANOVA will be button click away.

Reg. Var.Rep.Protein Means

1 1 1 ?????

1 1 2 ?????

1 1 3 ????? 1.82

1 2 1 ?????

1 2 2 ?????

1 2 3 ????? 1.37

1 3 1 ?????

1 3 2 ?????

1 3 3 ????? 1.67

2 1 1 ?????

2 1 2 ?????

2 1 3 ????? 1.36

2 2 1 ?????

2 2 2 ?????

2 2 3 ????? 1.18

2 3 1 ?????

2 3 2 ?????

2 3 3 ????? 1.26

3 4 1 ?????

3 4 2 ?????

3 4 3 ????? 1.04

Best regards

Elijah

1 Recommendation

5th May, 2012

Dr. Gajanan Vaishnav

YIPS

Analyze > General linear model > Univariate. Then select the dependent variable and the fixed factors (independent variables). You can select post hoc tests from the panel on right hand side. Then go to options and select homogeneity of variance test.

6th Jun, 2012

Shashi Ajit Chiplonkar

Jehangir Hospital

Thanks to all for your comments and suggestions.

6th Jun, 2012

Victor Mela O. Okoro

Federal University of Technology Owerri

Great solution. But SAS does a clearer and straight analysis when Analyst method is used.

6th Jun, 2012

Barbara Lee

Keiser Career College

Even simpler.. look it up in syntax, or type it into a syntax window: Manova protein by region (1,5) variety (1,10)/omeans.

That will give you main effects and interactions for region and variety, give you the means and standard deviationsfor each cell of the design, and univariate significance for each cell as well. This multivariate analysis of variance is unbiased about which independent variable might be most important. You can do the same thing in the GLM (general linear model) drop down menu, but the syntax for MANOVA is so simple, I usually just type it in as above. It really is just the one line. If you have a big enough sample for it to make sense, you can also do three way, fourway, etc. analysis of variance, not just two way. It is also easy to throw in covariates, just by addint " with covar". Categorical variables that you want to partial out of the analysis can be recoded to 1 variable for each category, using 1 for "yes" and 2 for "no", and used as "with" after the dependent variables, those effects will be removed from the model when the central hypotheses about the dependent variables are tested. Multivariate is the way to go. It is much closer to the real world, and allows testing for confounding variables and alternate explanations. I was able to make a very plausible case for cause and effect linkage regarding a program evaluation for which I could do only a post hoc design - the result was also validated with qualitative methods.Bottom line: MANOVA's and I have been friends since 1982, introduced by a very mathematical "real" statistician, so MANOVA and I know each other well, and have been collaborators many times since, with good results, and used in recent published papers (2009, 2010, and one now in ed after revisions in 2011).

1 Recommendation

7th Jul, 2012

Amina Kadiri

University of Oran 1 Ahmed Ben Bella

Hello

i am in the same situation i want to analyse my results by it's so difficult to know how entering data for spss

the experimental disgne is

3 genotypes of chickpea

2explants for each genotype

5 culture medium

to study the effect of these factors on callogenesis

each case is replicated 5 time

thank you

7th Jul, 2012

Sandor Schmikli

University Medical Center Utrecht

That's a three way Manova with five dependents. You will need the full model, taking into account all possible interactions.

You could also use the repeated design to exclude replication effects.

7th Jul, 2012

Amina Kadiri

University of Oran 1 Ahmed Ben Bella

Thank you for the answer but is it possible to guide me step by step

7th Jul, 2012

Sandor Schmikli

University Medical Center Utrecht

What SPSS version do you use?

7th Jul, 2012

Amina Kadiri

University of Oran 1 Ahmed Ben Bella

hello it's verry kind of you M Sandor Schmikli thank you very much i have Spss 17 Spss 20 and a trial version of Statistica 10 i am tring since a month ago to find the most simple and logical way but it is so difficulft

1st Jan, 2013

Amina Kadiri

University of Oran 1 Ahmed Ben Bella

thank you very much Mr Muhammad i used Statisca but with you link i'll compare with Spss thanks

1st Jan, 2013

Amina Kadiri

University of Oran 1 Ahmed Ben Bella

Dear Mr Muhammad, the problem with spss is in post hoc test eg. when you have factor1 factor 2 factor 3 and it is not possible to compare means of f1*f2*f3 or f1*f2 and give homogenous groups while with statistica it is

1st Jan, 2013

Carlos Jimenez-Gallardo

Universidad de La Frontera

analysis depends on the number of dependent variables

Please confirm, two variables, cereals and "places" (soil or climate conditions). What is the dependent variable or measurement ....?????

if a dependent variable univariate tests

Then check assumptions of normality, Shapiro-Wilk advise or Kolmogorov-Smirnov. Homoscedasticity analysis followed (for post hoc)

If multiple dependent variables multivariate tests

the normality assumption is tested through test of sphericity.

1st Jan, 2013

Amina Kadiri

University of Oran 1 Ahmed Ben Bella

One dependent factor and three independent

1st Jan, 2013

Carlos Jimenez-Gallardo

Universidad de La Frontera

Which is dependent factor?

1 Recommendation

1st Jan, 2013

Carlos Jimenez-Gallardo

Universidad de La Frontera

@amina SPSS version is not important,, How many replications?

1st Jan, 2013

Amina Kadiri

University of Oran 1 Ahmed Ben Bella

5 replications

factors1,2,3 indep and f 4dependent

1st Jan, 2013

Carlos Jimenez-Gallardo

Universidad de La Frontera

One or four dependet

1st Jan, 2013

Carlos Jimenez-Gallardo

Universidad de La Frontera

I send you a sample file, how to sort the data to SPSS-

• example data for spss.xlsx11.57 KB

1st Jan, 2013

Amina Kadiri

University of Oran 1 Ahmed Ben Bella

Thank you very much it is exactly my sample file with only one dependent factor

1st Jan, 2013

Amina Kadiri

University of Oran 1 Ahmed Ben Bella

Like this

• g art ang.xlsx13.32 KB

1st Jan, 2013

Amina Kadiri

University of Oran 1 Ahmed Ben Bella

Excuse me it is not the the good file

• example1.xlsx8.38 KB

1st Jan, 2013

Carlos Jimenez-Gallardo

Universidad de La Frontera

In examples i don`t see replications

1st Jan, 2013

Carlos Jimenez-Gallardo

Universidad de La Frontera

@amina. This is file for spss

• amina.xlsx8.71 KB

9th Sep, 2015

Ankit Agrawal

Acropolis Institute of Technology and Research

Please download video on youtube:  www.youtube.com/watch?v=6Hmauags3VA 

2 Recommendations

9th Sep, 2015

Shashi Ajit Chiplonkar

Jehangir Hospital

Thanks Ankit. It was interesting to watch the video though I got the answer before. Are there more such videos on other applications in SPSS?

10th Oct, 2016

Sakthivel Govindaraj

University of Madras

 Hello Diogo Pinho,

I am working in the field of stress physiology, I am comparing different parameters between Male animals Vs Female animals and within Male and female animals, each group contains three different time duration of stress. For this, I have to do two-way ANOVA in the SPSS (version 20) software. would you please explain to me how to do two-way for in this case.

Methods

I am giving stess to animal during the gestational period with three different trimester first, second and third trimester.

Example 1: for example i am calculating body weight gain during gestational period 1-10 days and 11-22 days gestational period for all  stress exposed groups and a control groups.

Example 2 : I am doing Neurobehavioural analysis (anxiety behaviour) for offspring both sex male and female of three stress exposed group and a control group.

how to do two-way ANOVA for these cases and how to enter the data in the datasheet. In these cases which i have to take as dependent variable and fixed factor.

Here is the sample data in the data sheet. please help me out

Thanks in advance for your valuable suggestion.  

• Two-way ANOVA sample.sav859 B

10th Oct, 2016

Victor Mela O. Okoro

Federal University of Technology Owerri

Dear Govindaraj,

I think the video from Ankit will go a long way.  But in your case, since each of the independent variable has other factors confounded in them, multivariate option of the General Linear Model will be the most appropriate.  It will help you capture all the effects you want regarding those factors of interest.  I hope it will help...

2nd Feb, 2017

Rajasekhar Chowdary Duddukur

Central Agricultural University

sir/madam

can anyone help me to analyse my Experimental data of 40 varieties of Blackgram, 3 replications for 13 characters  

My Research topic is "CHARACTER ASSOCIATION AND PATH ANALYSIS FOR SEED YIELD & ITS COMPONENTS IN BLACKGRAM (Vigna mungo (L.) Hepper)"

How can i execute GCV PCV heritability character association Genetic diversity cluster analysis & Path analysis by using IBM SPSS 

• BIOLOGICAL WEIGHT.sav3.09 KB

1 Recommendation

Can you help by adding an answer?

Answer