0
$\begingroup$

I need a little bit of help and confirmation that I have the right idea. I have some fake data of 8 tribes; within each tribe members work hard to gain food for their own tribe. No one can speak to these tribes, but people suspect that each tribe has one of the two strategies presented below for gaining food:

  1. Members of a tribe who travel farther away from the tribe's main location are given more food, so they face less of a chance of starving before coming back.

  2. Members of a tribe who travel far are given less food; that way if they are lost, there is less of a food loss to the tribe as a whole.

The variables (columns) I am working with are Tribe number, Distance from the tribe location when sample was taken (10, 20, 30, or 40 miles), Weight of each member that we are studying (related to the amount of food is given), height (taller people use energy more efficiently, and there is a strong positive correlation between weight and height in arbitrary units and inches), finally I have each observation categorized by height (group 1: 56–62in, group 2: 62–64...).

I want to find out if the tribes use different strategies, and also if there is a difference among the classes pf height. In addition I want to find out the strategies that are in use. I am having a hard time with how to classify each tribe as using either strategy 1 or 2. I was thinking of doing a one-way ANOVA to check if there is a difference in mean within each group based on distance. (In a particular tribe is there a difference in the mean of weight between those who were 10, 20 , 30, or 40 miles?) I don't know how to figure out if each colony uses a different strategy.

Finally, I want to build a linear model of mass on colony, distance, and height. I know how to build a model and run diagnostics. My concern here is, can I use distance as a categorical variable since its values are 10, 20, 30, or 40 miles?

Improve this question
$\endgroup$
3
  • 1
    $\begingroup$ Equating modeling building with ANOVA may mean that you should spend some time in background reading, especially related to how to model continuous variables without assuming linearity. $\endgroup$
    – Frank Harrell
    Commented Mar 29, 2014 at 20:13
  • $\begingroup$ Welcome to our site! This is a large set of questions. Although they are all on topic and interesting, there is too much to cover at once. To make this post answerable, please think over your situation and edit it to focus on one question. $\endgroup$
    – whuber ♦
    Commented Mar 29, 2014 at 20:17
  • $\begingroup$ Why is it fake data? Is it because this is for some sort of homework? If it is, the question should have the self-study tag. See stats.stackexchange.com/tags/self-study/info $\endgroup$
    – Patrick Coulombe
    Commented Mar 29, 2014 at 20:33

0

Reset to default

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.