topic Re: Confusing questions Tukey's Test and linear regression and quadratic models in SAS Programming

Confusing questions Tukey's Test and linear regression and quadratic models

syd1717 — Mon, 16 Nov 2020 04:04:56 GMT

Hello everyone.

I have come across questions that are somehow even more challenging for me to understand. The first question I am attaching my code that is giving me an output but I am not sure if it makes sense in terms of answering the questions. I do not know anything about Tukey's test.

The second question, I lost my code, but I didn't have much to begin with because I do not know much about regression models on SAS.

Please send some critique/help! Thank you in advance.

Re: Confusing questions Tukey's Test and linear regression and quadratic models

syd1717 — Mon, 16 Nov 2020 05:16:28 GMT

UPDATE: I think I may have figured out Tukey's Test for Question 1, I used a One Way ANOVA. Is that right? I am attaching my code. As for question 2, I am confused about the whole process besides inputting the data.

Re: Confusing questions Tukey's Test and linear regression and quadratic models

PaigeMiller — Mon, 16 Nov 2020 12:45:46 GMT

I (and many people) do not download and open attachments. Please include code in your message as text pasted into the window that appears when you click on the "running man" icon. Please include other parts of your document in your message as well.

Re: Confusing questions Tukey's Test and linear regression and quadratic models

syd1717 — Mon, 16 Nov 2020 16:21:51 GMT

Here is my code:

TITLE "Hypothesis Test";

data responses;

input circuittype$ responsetime@@;

cards;

1 9 1 12 1 10 1 8 1 15

2 20 2 21 2 23 2 20 2 30

3 6 3 15 3 8 3 16 3 7

;

run;

TITLE 'Data';

proc print data = responses (obs=15);

run;

proc means data = responses N MEAN MEDIAN VAR;

class circuittype;

var responsetime;

run;

TITLE 'One Way ANOVA'

Title;

ods noproctitle;

ods graphics / imagemap=on;

proc glm data=WORK.RESPONSES plots(only)=(boxplot diagnostics);

class circuittype;

model responsetime=circuittype;

means circuittype / hovtest=levene welch plots=none;

lsmeans circuittype / adjust=tukey pdiff alpha=.01 plots=(meanplot diffplot);

run;