Is Tukey the best way to test for differences in slopes after a mixed-effect ANCOVA?R, LME and Tukey Test produces error after sortingnegative coefficient of predictor in mixed-effects model produces positive slope in ggplot2Specify within-subjects and between-subjects ANOVA model using lme or lmer, as fixed-effectsGenerating similar estimates of interactions in afex, lsmeans, and lme4 packagesmodelling interaction terms in random effects and coding of daytime in growth models lme4Getting wrong p-values for Tukey test for one-way mixed effect ANOVATukey test only plot significant mean differencespost-hoc in r with more than one factorsTuckey correction for planned contrasts with emmeans and pairs() in RTukey test after LMM keeping contrasts
Informal question construction: "Anyone know what...", "Everyone finished?"
Is being an extrovert a necessary condition to be a manager?
One word for 'the thing that attracts me'?
How to tease a romance without a cat and mouse chase?
Is there a word for pant sleeves?
Why is a weak base more able to deprotonate a strong acid than a weak acid?
why "American-born", not "America-born"?
Is it normal to "extract a paper" from a master thesis?
Can diplomats be allowed on the flight deck of a commercial European airline?
Find this Unique UVC Palindrome ( ignoring signs and decimal) from Given Fractional Relationship
How can I reduce the size of matrix?
Computing elements of a 1000x60 matrix exhausts RAM
Way of refund if scammed?
What to call a small, open stone or cement reservoir that supplies fresh water from a spring or other natural source?
Which values for voltage divider
Why did Nick Fury not hesitate in blowing up the plane he thought was carrying a nuke?
Meaning of "half-crown enclosure"
Can a UK national work as a paid shop assistant in the USA?
Why the work done is positive when bringing 2 opposite charges together?
How do I write real-world stories separate from my country of origin?
Is a world with one country feeding everyone possible?
Island Perimeter
(For training purposes) Are there any openings with rook pawns that are more effective than others (and if so, what are they)?
What is the winged creature on the back of the Mordenkainen's Tome of Foes book?
Is Tukey the best way to test for differences in slopes after a mixed-effect ANCOVA?
R, LME and Tukey Test produces error after sortingnegative coefficient of predictor in mixed-effects model produces positive slope in ggplot2Specify within-subjects and between-subjects ANOVA model using lme or lmer, as fixed-effectsGenerating similar estimates of interactions in afex, lsmeans, and lme4 packagesmodelling interaction terms in random effects and coding of daytime in growth models lme4Getting wrong p-values for Tukey test for one-way mixed effect ANOVATukey test only plot significant mean differencespost-hoc in r with more than one factorsTuckey correction for planned contrasts with emmeans and pairs() in RTukey test after LMM keeping contrasts
.everyoneloves__top-leaderboard:empty,.everyoneloves__mid-leaderboard:empty,.everyoneloves__bot-mid-leaderboard:empty height:90px;width:728px;box-sizing:border-box;
I have a fairly simple fitted model that looks like:
fm <- lmer(Height~Site*HW+(1|Plot))
where Height and HW are continuous variables and Site and Plot are categorical. Site has 3 levels (A, B, C) I ran a Type II Wald Chi square and it showed that the interaction term was significant, which I am interpreting as that sites differed in their Height vs HW slopes:
> Anova(fm)
Analysis of Deviance Table (Type II Wald chisquare tests)
Response: Height
Chisq Df Pr(>Chisq)
Site 26.147 2 2.1e-06 ***
HW 91.089 1 < 2e-16 ***
Site:HW 13.775 2 0.00102 **
I'm interested in running a post-hoc to see which sites did and did not significantly differ in slope. I tried the following, but it doesn't look like a good match of what's going on visually when I plot the data. Is this the correct code?
leastsquare = lsmeans(fm,pairwise ~ Site:HW,adjust = "tukey")
leastsquare$contrasts
contrast estimate SE df t.ratio p.value
A - B 0.00206 0.0113 4.87 0.182 0.9819
A - C -0.04496 0.0101 4.88 -4.438 0.0163
B - C -0.04703 0.0113 4.87 -4.154 0.0212
So that's my first question: Is the above the correct way to test for slope differences among 3 levels of my categorical variable?
I am also interested in knowing how the means of the three site levels differ from one another (in addition to the slope question above). Can I just run a post-hoc tukey on Site? Is gives me warnings if I do this. Is there a better way to ask about main effects while taking into account the interaction?
Thank you in advance!
r
asked Mar 23 at 20:29
doingmybestdoingmybest
11
add a comment |
I have a fairly simple fitted model that looks like:
fm <- lmer(Height~Site*HW+(1|Plot))
where Height and HW are continuous variables and Site and Plot are categorical. Site has 3 levels (A, B, C) I ran a Type II Wald Chi square and it showed that the interaction term was significant, which I am interpreting as that sites differed in their Height vs HW slopes:
> Anova(fm)
Analysis of Deviance Table (Type II Wald chisquare tests)
Response: Height
Chisq Df Pr(>Chisq)
Site 26.147 2 2.1e-06 ***
HW 91.089 1 < 2e-16 ***
Site:HW 13.775 2 0.00102 **
I'm interested in running a post-hoc to see which sites did and did not significantly differ in slope. I tried the following, but it doesn't look like a good match of what's going on visually when I plot the data. Is this the correct code?
leastsquare = lsmeans(fm,pairwise ~ Site:HW,adjust = "tukey")
leastsquare$contrasts
contrast estimate SE df t.ratio p.value
A - B 0.00206 0.0113 4.87 0.182 0.9819
A - C -0.04496 0.0101 4.88 -4.438 0.0163
B - C -0.04703 0.0113 4.87 -4.154 0.0212
So that's my first question: Is the above the correct way to test for slope differences among 3 levels of my categorical variable?
I am also interested in knowing how the means of the three site levels differ from one another (in addition to the slope question above). Can I just run a post-hoc tukey on Site? Is gives me warnings if I do this. Is there a better way to ask about main effects while taking into account the interaction?
Thank you in advance!
r
asked Mar 23 at 20:29
doingmybestdoingmybest
11
add a comment |
I have a fairly simple fitted model that looks like:
fm <- lmer(Height~Site*HW+(1|Plot))
where Height and HW are continuous variables and Site and Plot are categorical. Site has 3 levels (A, B, C) I ran a Type II Wald Chi square and it showed that the interaction term was significant, which I am interpreting as that sites differed in their Height vs HW slopes:
> Anova(fm)
Analysis of Deviance Table (Type II Wald chisquare tests)
Response: Height
Chisq Df Pr(>Chisq)
Site 26.147 2 2.1e-06 ***
HW 91.089 1 < 2e-16 ***
Site:HW 13.775 2 0.00102 **
I'm interested in running a post-hoc to see which sites did and did not significantly differ in slope. I tried the following, but it doesn't look like a good match of what's going on visually when I plot the data. Is this the correct code?
leastsquare = lsmeans(fm,pairwise ~ Site:HW,adjust = "tukey")
leastsquare$contrasts
contrast estimate SE df t.ratio p.value
A - B 0.00206 0.0113 4.87 0.182 0.9819
A - C -0.04496 0.0101 4.88 -4.438 0.0163
B - C -0.04703 0.0113 4.87 -4.154 0.0212
So that's my first question: Is the above the correct way to test for slope differences among 3 levels of my categorical variable?
I am also interested in knowing how the means of the three site levels differ from one another (in addition to the slope question above). Can I just run a post-hoc tukey on Site? Is gives me warnings if I do this. Is there a better way to ask about main effects while taking into account the interaction?
Thank you in advance!
r
asked Mar 23 at 20:29
doingmybestdoingmybest
11
I have a fairly simple fitted model that looks like:
fm <- lmer(Height~Site*HW+(1|Plot))
where Height and HW are continuous variables and Site and Plot are categorical. Site has 3 levels (A, B, C) I ran a Type II Wald Chi square and it showed that the interaction term was significant, which I am interpreting as that sites differed in their Height vs HW slopes:
> Anova(fm)
Analysis of Deviance Table (Type II Wald chisquare tests)
Response: Height
Chisq Df Pr(>Chisq)
Site 26.147 2 2.1e-06 ***
HW 91.089 1 < 2e-16 ***
Site:HW 13.775 2 0.00102 **
I'm interested in running a post-hoc to see which sites did and did not significantly differ in slope. I tried the following, but it doesn't look like a good match of what's going on visually when I plot the data. Is this the correct code?
leastsquare = lsmeans(fm,pairwise ~ Site:HW,adjust = "tukey")
leastsquare$contrasts
contrast estimate SE df t.ratio p.value
A - B 0.00206 0.0113 4.87 0.182 0.9819
A - C -0.04496 0.0101 4.88 -4.438 0.0163
B - C -0.04703 0.0113 4.87 -4.154 0.0212
So that's my first question: Is the above the correct way to test for slope differences among 3 levels of my categorical variable?
I am also interested in knowing how the means of the three site levels differ from one another (in addition to the slope question above). Can I just run a post-hoc tukey on Site? Is gives me warnings if I do this. Is there a better way to ask about main effects while taking into account the interaction?
Thank you in advance!
r
r
asked Mar 23 at 20:29
doingmybestdoingmybest
11
asked Mar 23 at 20:29
doingmybestdoingmybest
11
edited Mar 23 at 20:41
doingmybest
asked Mar 23 at 20:29
doingmybestdoingmybest
11
asked Mar 23 at 20:29
doingmybestdoingmybest
11
asked Mar 23 at 20:29
doingmybestdoingmybest
11
11
add a comment |
add a comment |
1 Answer
1
active
oldest
votes
That’s because the code shown compares means, not slopes. Look at the emtrends (or lstrends) function. Both are documented in the emmeans package.
emt = emtrends(fm, “Site”, var = HW)
emt # estimated slopes
pairs(emt) # pairwise comparisons
answered Mar 24 at 17:32
rvlrvl
2,1322613
add a comment |
Your Answer
StackExchange.ifUsing("editor", function ()
StackExchange.using("externalEditor", function ()
StackExchange.using("snippets", function ()
StackExchange.snippets.init();
);
);
, "code-snippets");
StackExchange.ready(function()
var channelOptions =
tags: "".split(" "),
id: "1"
;
initTagRenderer("".split(" "), "".split(" "), channelOptions);
StackExchange.using("externalEditor", function()
// Have to fire editor after snippets, if snippets enabled
if (StackExchange.settings.snippets.snippetsEnabled)
StackExchange.using("snippets", function()
createEditor();
);
else
createEditor();
);
function createEditor()
StackExchange.prepareEditor(
heartbeatType: 'answer',
autoActivateHeartbeat: false,
convertImagesToLinks: true,
noModals: true,
showLowRepImageUploadWarning: true,
reputationToPostImages: 10,
bindNavPrevention: true,
postfix: "",
imageUploader:
brandingHtml: "Powered by u003ca class="icon-imgur-white" href="https://imgur.com/"u003eu003c/au003e",
contentPolicyHtml: "User contributions licensed under u003ca href="https://creativecommons.org/licenses/by-sa/3.0/"u003ecc by-sa 3.0 with attribution requiredu003c/au003e u003ca href="https://stackoverflow.com/legal/content-policy"u003e(content policy)u003c/au003e",
allowUrls: true
,
onDemand: true,
discardSelector: ".discard-answer"
,immediatelyShowMarkdownHelp:true
);
);
Sign up or log in
StackExchange.ready(function ()
StackExchange.helpers.onClickDraftSave('#login-link');
);
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
StackExchange.ready(
function ()
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fstackoverflow.com%2fquestions%2f55318075%2fis-tukey-the-best-way-to-test-for-differences-in-slopes-after-a-mixed-effect-anc%23new-answer', 'question_page');
);
Post as a guest
Required, but never shown
1 Answer
1
active
oldest
votes
1 Answer
1
active
oldest
votes
active
oldest
votes
active
oldest
votes
That’s because the code shown compares means, not slopes. Look at the emtrends (or lstrends) function. Both are documented in the emmeans package.
emt = emtrends(fm, “Site”, var = HW)
emt # estimated slopes
pairs(emt) # pairwise comparisons
answered Mar 24 at 17:32
rvlrvl
2,1322613
add a comment |
That’s because the code shown compares means, not slopes. Look at the emtrends (or lstrends) function. Both are documented in the emmeans package.
emt = emtrends(fm, “Site”, var = HW)
emt # estimated slopes
pairs(emt) # pairwise comparisons
answered Mar 24 at 17:32
rvlrvl
2,1322613
add a comment |
That’s because the code shown compares means, not slopes. Look at the emtrends (or lstrends) function. Both are documented in the emmeans package.
emt = emtrends(fm, “Site”, var = HW)
emt # estimated slopes
pairs(emt) # pairwise comparisons
answered Mar 24 at 17:32
rvlrvl
2,1322613
That’s because the code shown compares means, not slopes. Look at the emtrends (or lstrends) function. Both are documented in the emmeans package.
emt = emtrends(fm, “Site”, var = HW)
emt # estimated slopes
pairs(emt) # pairwise comparisons
answered Mar 24 at 17:32
rvlrvl
2,1322613
edited Mar 25 at 20:38
answered Mar 24 at 17:32
rvlrvl
2,1322613
answered Mar 24 at 17:32
rvlrvl
2,1322613
answered Mar 24 at 17:32
rvlrvl
2,1322613
2,1322613
add a comment |
add a comment |
Thanks for contributing an answer to Stack Overflow!
- Please be sure to answer the question. Provide details and share your research!
But avoid …
- Asking for help, clarification, or responding to other answers.
- Making statements based on opinion; back them up with references or personal experience.
To learn more, see our tips on writing great answers.
Sign up or log in
StackExchange.ready(function ()
StackExchange.helpers.onClickDraftSave('#login-link');
);
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
StackExchange.ready(
function ()
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fstackoverflow.com%2fquestions%2f55318075%2fis-tukey-the-best-way-to-test-for-differences-in-slopes-after-a-mixed-effect-anc%23new-answer', 'question_page');
);
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function ()
StackExchange.helpers.onClickDraftSave('#login-link');
);
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function ()
StackExchange.helpers.onClickDraftSave('#login-link');
);
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function ()
StackExchange.helpers.onClickDraftSave('#login-link');
);
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
