Closed cjvanlissa closed 5 years ago
I agree this is a good idea
related to #118
Is there a straightforward way to do something like this using base (especially for spread()
:
library(tidyLPA)
#> tidyLPA has received a major update, with a much easier workflow and improved functionality. However, you might have to update old syntax to account for the new workflow. See vignette('introduction-to-major-changes') for details!
#>
#> Mplus is not installed. Use only package = 'mclust' when calling estimate_profiles().
library(dplyr)
#>
#> Attaching package: 'dplyr'
#> The following objects are masked from 'package:stats':
#>
#> filter, lag
#> The following objects are masked from 'package:base':
#>
#> intersect, setdiff, setequal, union
library(tidyr)
m <- pisaUSA15[1:100, ] %>%
select(broad_interest, enjoyment, self_efficacy) %>%
single_imputation() %>%
estimate_profiles(3)
get_data(m) %>%
mutate(Class_prob = paste0("Class_", Class_prob)) %>%
spread(Class_prob, Probability)
#> # A tibble: 100 x 10
#> model_number classes_number broad_interest enjoyment self_efficacy Class
#> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1 1 3 0.853 2.4 1.57 3
#> 2 1 3 1 1 4 2
#> 3 1 3 1 1.2 3.5 2
#> 4 1 3 1 2 2 3
#> 5 1 3 1 2 3 3
#> 6 1 3 1 2.2 2.29 3
#> 7 1 3 1 2.6 1.86 3
#> 8 1 3 1 2.6 2.38 3
#> 9 1 3 1 2.8 2.62 3
#> 10 1 3 1 4 1.5 1
#> # … with 90 more rows, and 4 more variables: id <int>, Class_1 <dbl>,
#> # Class_2 <dbl>, Class_3 <dbl>
Created on 2019-07-17 by the reprex package (v0.2.1)
See my pull request: get_data
returns data in wide format when applied to an object of class tidyProfile
(one element of a tidyLPA object), or when applied to a tidyLPA
object of length one. It returns long format when applied to a tidyLPA
object containing multiple tidyProfile
analyses (because then the wide format does not make sense).
Thanks sir!
From: C. J. van Lissa notifications@github.com Sent: Wednesday, July 17, 2019 3:21 PM To: data-edu/tidyLPA Cc: Joshua Rosenberg; Comment Subject: Re: [data-edu/tidyLPA] get_data should return data in wide format (#117)
See my pull request: get_data returns data in wide format when applied to an object of class tidyProfile (one element of a tidyLPA object), or when applied to a tidyLPA object of length one. It returns long format when applied to a tidyLPA object containing multiple tidyProfile analyses (because then the wide format does not make sense).
— You are receiving this because you commented. Reply to this email directly, view it on GitHubhttps://urldefense.proofpoint.com/v2/url?u=https-3A__github.com_data-2Dedu_tidyLPA_issues_117-3Femail-5Fsource-3Dnotifications-26email-5Ftoken-3DABDCD5WJWOTI763FNR6RFJTP75WMXA5CNFSM4HQLRCO2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOD2GKC3A-23issuecomment-2D512532844&d=DwMCaQ&c=nE__W8dFE-shTxStwXtp0A&r=qjThNNzo58j8Je1aLG3p-w&m=qsLn8HXVhctcK8Ba0jgk5--jrZKk7tcF0yUY9tdyIZY&s=q72QwjYY2iohG2Njcmo2xW5WkJmuaauMaKicE4QV36k&e=, or mute the threadhttps://urldefense.proofpoint.com/v2/url?u=https-3A__github.com_notifications_unsubscribe-2Dauth_ABDCD5XCQRNJSVECLONM2GLP75WMXANCNFSM4HQLRCOQ&d=DwMCaQ&c=nE__W8dFE-shTxStwXtp0A&r=qjThNNzo58j8Je1aLG3p-w&m=qsLn8HXVhctcK8Ba0jgk5--jrZKk7tcF0yUY9tdyIZY&s=Tyk3qMAazvTqDtvi5H1ZPrsZ5GUDqvtbIDhitvm7gF0&e=.
This was addressed in #122
Can tidyLPA be used with repeated measures and if so, then should data be in long or wide format? Thanks!
@benjaminwnelson yes it can, but you can't fit a longitudinal model to the data. The data should be in wide format, multilevel mixture models are not available.
Thank you! Would you advise that after pulling out LPA groups, you could then join that factor variable with a long dataset in order to do MLM modeling with the LPA group membership?
These are substantive questions; I would recommend finding a statistical collaborator who can help out with this. But briefly: The function ?get_data does what you're asking, but this approach ignores classification uncertainty. If entropy is > .90, the resulting bias may be negligible. In other cases, you will need to conduct 3-step analysis, which uses the classification probability matrix (see the literature).
Yes, will definitely consult on this. Thank you!
Or create function that merges LPA results with original data file