# What kind of ANOVA is this?

#### eb11

##### New Member
Ok...I know this may be a silly question but I am extremely bad at stats.

I have an experiment with two IVs and a control group. ie. one group must perform a task after seeing someone complete it before them, another group must perform a task after watching someone attempt (and not finish) the task prior to them, and the control group who simply complete the task without watching anything. All my googling has told me is that there are two factors (two independent variables). So I know this is a _ x _ ANOVA, but I am not sure what. Any help would be GREATLY appreciated!!

#### Karabiner

##### TS Contributor
It is a oneway analysis of variance (if you really
had 2 factors, then you'd have to have at least
4 groups).

With kind regards

K.

#### noetsi

##### No cake for spunky
I think the original poster is confusing variables, different factors, with the levels of those factors. As Karabiner is pointing out you only have one thing you are testing, you just have 3 levels of that factor.

If you thought gender and age independently effected your DV than you would have two factors, and thus two way ANOVA. If you thought that age, split into three categories, was the only major impact than you have one way ANOVA (with a factor with 3 categories or levels).

When you have more than two factors you have a factorial design with specific nomeclature for the number of variables and levels of each. This may be what you are thinking about.

#### GretaGarbo

##### Human
Also, there might be a lack of independence if group A is watching group B and if B is "successful" then A also might be so.

Second: if they have all been attending in a specific group at the same time then the sample size is just one (1).
So the traditional question remains: How was the experiment done? Randomized? sample size? measured variable? distribution of that?

#### eb11

##### New Member
Hi, sorry if my original post wasn't clear...but I have literally been given a passage about an experiment, and the exercise is to correct the results section we've been given for it, and what tests you'd carry out. This is obviously very hard when you're so rusty on stats!

Basically we are looking at the task performance of people after having watched someone complete the task before them, watch someone attempt (and not finish) the task before them, or after having watched no one at all (control). I have been given ANOVA results to say that there was a significant effect of task status in that Pps who watched someone perform the task before them, did worse on their own task than those who observed someone just attempt the task before them.

So I know that an ANOVA is used..but I don't know what type and I would like if someone could explain to me which you'd use and the reasons (to help me understand), and then because it is significant.....what tests do I then carry out!? Thankyou to anyone who can help!!

#### noetsi

##### No cake for spunky
Given the information you have it looks like a one way ANOVA was carried out. They would have used it because other types of ANOVA such as repeated measures (which measures change over time) or factorial ANOVA (which is simply multiple predictors) were not needed for their research question. Commonly you will get an F test from that which tells you if it is signficant when you do this. It is common practice after that to conduct follow up post-hoc statistical tests which tells you what level varied, and how much, which the simple design you listed does not. If they had suspected a specific difference in levels (say the control group varied from those who watched the task) they could have generated pre-planned contrasts instead of ad hoc tests.