# I am getting asterisks in minitab for the p--value when i try to do a Two-Way ANOVA

I am trying to do a Two-Way ANOVA. I did what i think I need to do to have the p-value show up on the screen (after i click the button that starts the calculation) but, all i see next to the P-value and F-value is asterisks.

When i googled the problem I found the following answer on the website where you can download minitab. "The asterisks represent missing values that cannot be calculated because the model is saturated and there are not enough degrees of freedom for error." Because of this i added a couple more values but im still getting asterisk instead of what the p-value and f-value are equal to.

What specific things should I do to my data to get Minitab (a software that does statistical calculations) to calculate and show the p value, f value etc?

My data is the number of video game soccer games I won that lasted different lengths. Half the games were only seen by me. The other half of the games were seen by me and someone else. The goal of the assignment is to see if there is an effect and/or interaction between how long a video game soccer game is, and how many games I win, as well to find out if there is an effect and/or interaction between how many games I win that are seen by just me and games that are seen by me and somebody else.

Let me know if i should and how i should set this data up differently so that i get a F-value and a p-value. There is a link to a picture of the data below. (I don't have enough points to post a picture next to the text so it put a link to the picture.) What specific things should I do to my data to get Minitab (a software that does statistical calculations) to calculate and show the p value, f value etc.

My data is the number of video game soccer games I won that lasted different lengths. Here is a picture of the way i have set the data up

Apparently you are trying to run a two-factor ANOVA, with four levels of factor Time (12, 20, 30, and 40) and two levels of factor Watchers (0 and 1). In that case, you have one observation per cell.

I put the data into a Minitab worksheet as follows:

MTB > print c1-c3

Data Display

Row  Time  Watchers  Won
1    12         0    2
2    12         1    3
3    20         0    4
4    20         1    4
5    30         0    3
6    30         1    2
7    40         0    3
8    40         1    3


Put into tabular form you have:

 MTB > table c2 c1;
SUBC> data c3.

Tabulated statistics: Watchers, Time

Rows: Watchers   Columns: Time

12  20  30  40

0     2   4   3   3

1     3   4   2   3

Cell Contents:  Won  :  DATA


This design supports a two-way model $without$ interaction. In order to have a term in the model for interaction, you would need to have two or more replications in each cell of the table. A two-way ANOVA procedure without interaction gives the following ANOVA table and P-values.

 MTB > anova c3 = c1 c2

ANOVA: Won versus Time, Watchers

Factor    Type   Levels  Values
Time      fixed       4  12, 20, 30, 40
Watchers  fixed       2  0, 1

Analysis of Variance for Won

Source    DF      SS      MS     F      P
Time       3  3.0000  1.0000  3.00  0.196
Watchers   1  0.0000  0.0000  0.00  1.000
Error      3  1.0000  0.3333
Total      7  4.0000


It appears that neither factor is significant at any reasonable level of significance. (Notice, in particular, that the total number Won was the same (12) with and without someone watching.)