Add workflows for tests and R CMD check (#262)

* inital-attempt for running workflows * add Encoding: UTF-8 in DESCRIPTION * replace expect_is() since it is deprecated * dont run tests in check * fix partial matches * vuongtest -> nonnest2::vuongtest * expect_equal(tol=...) -> expect_equal(tolerance=...) * fix groupName partial match * unamed arg -> tolerance=... * tol -> tolerance in test-10-extras.R * dont trigger tests automatically on push to main
philchalmers · Nov 11, 2024 · 3c50fb7 · 3c50fb7
1 parent dbf9ad3
commit 3c50fb7
Show file tree

Hide file tree

Showing 44 changed files with 242 additions and 193 deletions.
diff --git a/.github/workflows/checks.yml b/.github/workflows/checks.yml
@@ -0,0 +1,23 @@
+name: R-CMD-CHECK
+
+on:
+  push:
+    branches: [main, master]
+  pull_request:
+    branches: [main, master]
+  workflow_dispatch:
+
+jobs:
+  R-CMD-CHECK:
+    runs-on: ubuntu-latest
+
+    container:
+      image: ghcr.io/kss2k/container-mirt:latest
+
+    steps:
+      - name: Check out the repository
+        uses: actions/checkout@v4
+
+      - name: R CMD CHECK
+        run: |
+          Rscript -e 'devtools::check(error_on = "warning")'
diff --git a/.github/workflows/tests.yml b/.github/workflows/tests.yml
@@ -0,0 +1,23 @@
+name: Tests (Ubuntu-latest)
+
+on:
+  # push:
+  #   branches: [main, master]
+  pull_request:
+    branches: [main, master]
+  workflow_dispatch:
+
+jobs:
+  Run-Tests:
+    runs-on: ubuntu-latest
+
+    container:
+      image: ghcr.io/kss2k/container-mirt:latest
+
+    steps:
+      - name: Check out the repository
+        uses: actions/checkout@v4
+
+      - name: Run tests
+        run: |
+          Rscript -e 'options(warnPartialMatchDollar = TRUE); Sys.setenv("_R_CHECK_LENGTH_1_CONDITION_" = "TRUE", "_R_CHECK_LENGTH_1_LOGIC2_" = "TRUE"); devtools::test(stop_on_failure = TRUE)'
diff --git a/DESCRIPTION b/DESCRIPTION
@@ -56,14 +56,17 @@ Suggests:
     nloptr,
     sirt,
     plink,
-    mirtCAT
+    mirtCAT,
+    testthat (>= 3.0.0)
 ByteCompile: yes
 LazyLoad: yes
 LazyData: yes
 LinkingTo: Rcpp, RcppArmadillo
 License: GPL (>= 3)
+Encoding: UTF-8
 Repository: CRAN
 Maintainer: Phil Chalmers <rphilip.chalmers@gmail.com>
 URL: https://philchalmers.github.io/mirt/, https://github.com/philchalmers/mirt/wiki, https://groups.google.com/forum/#!forum/mirt-package
 BugReports: https://github.com/philchalmers/mirt/issues?state=open
 RoxygenNote: 7.3.2
+Config/testthat/edition: 3
diff --git a/R/03-estimation.R b/R/03-estimation.R
@@ -867,8 +867,8 @@ ESTIMATION <- function(data, model, group, itemtype = NULL, guess = 0, upper = 1
                             Consider changing the starting values', call.=FALSE)
                 dontrun <- TRUE
             }
-            lengthsplit <- do.call(c, lapply(strsplit(names(ESTIMATE$correct), 'COV_'), length))
-            lengthsplit <- lengthsplit + do.call(c, lapply(strsplit(names(ESTIMATE$correct), 'MEAN_'), length))
+            lengthsplit <- do.call(c, lapply(strsplit(names(ESTIMATE$correction), 'COV_'), length))
+            lengthsplit <- lengthsplit + do.call(c, lapply(strsplit(names(ESTIMATE$correction), 'MEAN_'), length))
             is.latent <- lengthsplit > 2L
             if(!dontrun){
                 if(ESTIMATE$cycles <= 10L)
@@ -1016,7 +1016,7 @@ ESTIMATION <- function(data, model, group, itemtype = NULL, guess = 0, upper = 1
                 .mirtClusterEnv$ncores <- 1L
             }
             for(g in seq_len(Data$ngroups)){
-                cmods[[g]]@Data <- list(data=Data$data[Data$group == Data$groupName[g], ],
+                cmods[[g]]@Data <- list(data=Data$data[Data$group == Data$groupNames[g], ],
                                         fulldata=Data$fulldata[[g]], tabdata=Data$tabdata,
                                         Freq=list(Data$Freq[[g]]), K=Data$K)
                 cmods[[g]] <- calcLogLik(cmods[[g]], opts$draws, G2 = 'return',
@@ -1166,7 +1166,7 @@ ESTIMATION <- function(data, model, group, itemtype = NULL, guess = 0, upper = 1
             } else {
                 if(Options$exploratory){
                     FF <- F %*% t(F)
-                    V <- eigen(FF)$vector[ ,1L:nfact]
+                    V <- eigen(FF)$vectors[ ,1L:nfact]
                     L <- eigen(FF)$values[1L:nfact]
                     if (nfact == 1L) F <- as.matrix(V * sqrt(L))
                     else F <- V %*% sqrt(diag(L))

diff --git a/R/05-model.elements.R b/R/05-model.elements.R
@@ -98,7 +98,7 @@ model.elements <- function(model, factorNames, itemtype, nfactNames, nfact, J, K
 
     if(exploratory){
         Rpoly <- cormod(data, K, guess)
-        loads <- eigen(Rpoly)$vector[,seq_len(nfact), drop = FALSE]
+        loads <- eigen(Rpoly)$vectors[,seq_len(nfact), drop = FALSE]
         u <- 1 - rowSums(loads^2)
         u[u < .001 ] <- .2
         cs <- sqrt(u)

diff --git a/R/EMstep.group.R b/R/EMstep.group.R
@@ -150,7 +150,7 @@ EM.group <- function(pars, constrain, Ls, Data, PrepList, list, Theta, DERIV, so
         tmp <- tmp[lower.tri(tmp, TRUE)]
         tmpmat <- matrix(0, ns, 2L)
         for(i in seq_len(ns))
-            tmpmat[i, ] <- c(gp$gmean[np + i], gp$gcov[np+i, np+i])
+            tmpmat[i, ] <- c(gp$gmeans[np + i], gp$gcov[np+i, np+i])
         for(g in seq_len(ngroups)){
             pars[[g]][[J+1L]]@bindex <- as.integer(c(gp$gmeans[seq_len(np)], tmp))
             pars[[g]][[J+1L]]@sindex = tmpmat

diff --git a/R/M2.R b/R/M2.R
@@ -381,7 +381,7 @@ M2 <- function(obj, type="M2*", calcNull = TRUE, quadpts = NULL, theta_lim = c(-
                 pars[[g]]@Internals$Prior <- list(obj@Internals$Prior[[g]])
                 pars[[g]]@Model$Theta <- obj@Model$Theta
             }
-            pars[[g]]@Data <- list(data=obj@Data$data[obj@Data$group == obj@Data$groupName[g], ],
+            pars[[g]]@Data <- list(data=obj@Data$data[obj@Data$group == obj@Data$groupNames[g], ],
                                    mins=obj@Data$mins, K=obj@Data$K,
                                    fulldata=list(obj@Data$fulldata[[g]]))
             if(is(obj, 'MixtureClass')) pars[[g]]@Data$data <- extract.mirt(obj, 'data')

diff --git a/R/MultipleGroup-methods.R b/R/MultipleGroup-methods.R
@@ -341,7 +341,7 @@ setMethod(
         for(g in 1L:length(ret)){
             cmod <- object@ParObjects$pars[[g]]
             cmod@Data <- object@Data
-            cmod@Data$data <- object@Data$data[object@Data$group == object@Data$groupName[g], ]
+            cmod@Data$data <- object@Data$data[object@Data$group == object@Data$groupNames[g], ]
             cmod@Data$Freq[[1L]] <- cmod@Data$Freq[[g]]
             cmod@Options$quadpts <- object@Options$quadpts
             cmod@Internals$bfactor <- object@Internals$bfactor

diff --git a/R/fscores.internal.R b/R/fscores.internal.R
@@ -864,7 +864,7 @@ EAPsum <- function(x, full.scores = FALSE, full.scores.SE = FALSE,
     factorNames <- extract.mirt(x, 'factorNames')
     colnames(thetas) <- factorNames[!grepl('\\(',factorNames)]
     colnames(SEthetas) <- paste0('SE_', colnames(thetas))
-    ret <- data.frame(Sum.Scores=Sum.Scores + sum(x@Data$min), thetas, SEthetas)
+    ret <- data.frame(Sum.Scores=Sum.Scores + sum(x@Data$mins), thetas, SEthetas)
     rownames(ret) <- ret$Sum.Scores
     if(full.scores){
         dat <- x@Data$data
@@ -1082,4 +1082,4 @@ EAP_classify <- function(ID, log_itemtrace, tabdata, W, nclass){
     }
     probs <- matrix(expLW / nc, ncol=nclass)
     colSums(probs)
-}
+}
diff --git a/R/utils.R b/R/utils.R
@@ -95,7 +95,7 @@ thetaStack <- function(theta, nclass){
 #' }
 secondOrderTest <- function(mat, ..., method = 'eigen'){
     if(method == 'eigen'){
-        evs <- eigen(mat, ...)$value
+        evs <- eigen(mat, ...)$values
         ret <- all(!sapply(evs, function(x) isTRUE(all.equal(x, 0))) & evs > 0)
     } else if(method == 'chol'){
         chl <- try(chol(mat, ...), silent = TRUE)
@@ -2428,7 +2428,7 @@ MGC2SC <- function(x, which){
     }
     tmp@Data <- x@Data
     tmp@Data$completely_missing <- integer(0L)
-    tmp@Data$data <- tmp@Data$data[tmp@Data$group == tmp@Data$groupName[which], , drop=FALSE]
+    tmp@Data$data <- tmp@Data$data[tmp@Data$group == tmp@Data$groupNames[which], , drop=FALSE]
     tmp@Data$rowID <- 1L:nrow(tmp@Data$data)
     tmp@Data$Freq[[1L]] <- tmp@Data$Freq[[which]]
     tmp@Data$fulldata[[1L]] <- x@Data$fulldata[[which]]

diff --git a/tests/tests/test-00-basics.R → tests/testthat/test-00-basics.R b/tests/tests/test-00-basics.R → tests/testthat/test-00-basics.R
@@ -1,4 +1,4 @@
-context('basics')
+expect_class <- function(x, class) expect_true(inherits(x, class))
 
 test_that('basics', {
     group <- rep(c('G1', 'G2'), each=nrow(Science)/2)
@@ -19,7 +19,7 @@ test_that('basics', {
     tscore <- expected.test(mod, Theta)
     expect_equal(tscore, c(7.935196,9.270222,10.53705,11.69166,12.7803,13.83405,14.73532), tolerance=1e-6)
     IG <- itemGAM(Science[group == 'G1',1], theta_se[,1, drop=FALSE])
-    expect_is(IG, 'itemGAM')
+    expect_class(IG, 'itemGAM')
 
     info <- iteminfo(item1, Theta)
     expect_equal(as.vector(info), c(0.2031261,0.2118485,0.2069816,0.187747,0.1793368,0.1902587,0.1761818),

diff --git a/tests/tests/test-01-mirtOne.R → tests/testthat/test-01-mirtOne.R b/tests/tests/test-01-mirtOne.R → tests/testthat/test-01-mirtOne.R
@@ -1,17 +1,17 @@
-context('mirtOne')
+expect_class <- function(x, class) expect_true(inherits(x, class))
 
 test_that('dich', {
     data <- expand.table(LSAT7)
     mod1 <- mirt(data, 1, verbose=FALSE)
-    expect_is(mod1, 'SingleGroupClass')
+    expect_class(mod1, 'SingleGroupClass')
     expect_equal(extract.mirt(mod1, 'df'), 21)
     cfs <- as.numeric(do.call(c, coef(mod1)))
     expect_equal(cfs, c(0.988, 1.8561, 0, 1, 1.081, 0.808, 0, 1, 1.706, 1.8043, 0, 1, 0.7651, 0.486, 0, 1, 0.7358, 1.8545, 0, 1, 0, 1),
                  tolerance = 1e-2)
     sv <- mod2values(mod1)
     sv$est <- FALSE
     moddummy <- mirt(data, 1, pars= sv, verbose=FALSE)
-    expect_is(moddummy, 'SingleGroupClass')
+    expect_class(moddummy, 'SingleGroupClass')
     sv2 <- mod2values(moddummy)
     expect_equal(sv$value, sv2$value)
     modm1 <- mirt(data, 1, SE = TRUE, SE.type = 'SEM', verbose=FALSE)
@@ -20,17 +20,17 @@ test_that('dich', {
     expect_equal(extract.mirt(modm1, 'condnum'), 30.12751, tolerance = 1e-4)
     expect_equal(cfs, c(0.9876, 0.6367, 1.3384, 1.8559, 1.5978, 2.1139, 0, NA, NA, 1, NA, NA, 1.0808, 0.7604, 1.4013, 0.808, 0.6335, 0.9825, 0, NA, NA, 1, NA, NA, 1.7075, 1.0868, 2.3281, 1.8052, 1.4028, 2.2076, 0, NA, NA, 1, NA, NA, 0.765, 0.5065, 1.0235, 0.486, 0.3114, 0.6606, 0, NA, NA, 1, NA, NA, 0.7357, 0.4246, 1.0467, 1.8545, 1.6332, 2.0757, 0, NA, NA, 1, NA, NA, 0, NA, NA, 1, NA, NA),
                  tolerance = 1e-2)
-    expect_is(modm1, 'SingleGroupClass')
+    expect_class(modm1, 'SingleGroupClass')
     modm2 <- mirt(data, 1, SE = TRUE, SE.type = 'Richardson', verbose=FALSE)
     cfs <- as.numeric(do.call(c, coef(modm2)))
     expect_equal(extract.mirt(modm2, 'condnum'), 30.24068, tolerance = 1e-3)
     expect_equal(cfs, c(0.988, 0.6406, 1.3354, 1.8561, 1.5984, 2.1138, 0, NA, NA, 1, NA, NA, 1.081, 0.7501, 1.4119, 0.808, 0.6291, 0.9869, 0, NA, NA, 1, NA, NA, 1.706, 1.0779, 2.334, 1.8043, 1.4036, 2.205, 0, NA, NA, 1, NA, NA, 0.7651, 0.5022, 1.028, 0.486, 0.3392, 0.6328, 0, NA, NA, 1, NA, NA, 0.7358, 0.4395, 1.032, 1.8545, 1.6302, 2.0787, 0, NA, NA, 1, NA, NA, 0, NA, NA, 1, NA, NA),
                  tolerance = 1e-2)
-    expect_is(modm2, 'SingleGroupClass')
+    expect_class(modm2, 'SingleGroupClass')
     modm2b <- mirt(data, 1, SE = TRUE, SE.type = 'Fisher', verbose=FALSE)
     expect_equal(extract.mirt(modm2b, 'condnum'), 29.0323, tolerance = 1e-3)
     modm3 <- mirt(data, 1, itemtype = 'Rasch', verbose=FALSE, SE=TRUE)
-    expect_is(modm3, 'SingleGroupClass')
+    expect_class(modm3, 'SingleGroupClass')
     expect_equal(extract.mirt(modm3, 'df'), 25)
     expect_equal(extract.mirt(modm3, 'condnum'), 4.488772, tolerance = 1e-4)
     LG <- lagrange(modm3, parnum = list(1, 5))
@@ -39,7 +39,7 @@ test_that('dich', {
     expect_equal(LG2$X2, 0.4816444, tolerance = 1e-4)
     dat <- expand.table(LSAT6)
     modm3 <- mirt(dat, 1, itemtype = 'Rasch', SE = TRUE, SE.type = 'SEM', verbose=FALSE)
-    expect_is(modm3, 'SingleGroupClass')
+    expect_class(modm3, 'SingleGroupClass')
     cfs <- as.numeric(do.call(c, coef(modm3)))
     expect_equal(cfs, c(1,NA,NA,2.73,2.478,2.983,0,NA,NA,1,NA,NA,1,NA,NA,0.999,0.845,1.152,0,NA,NA,1,NA,NA,1,NA,NA,0.24,0.1,0.38,0,NA,NA,1,NA,NA,1,NA,NA,1.306,1.143,1.47,0,NA,NA,1,NA,NA,1,NA,NA,2.099,1.896,2.303,0,NA,NA,1,NA,NA,0,NA,NA,0.57,0.371,0.77),
                  tolerance = 1e-2)
@@ -58,18 +58,18 @@ test_that('dich', {
     svalues <- mirt(data, 1, pars = 'values', verbose=FALSE)
     svalues[22, 'value'] <- 2
     modm5 <- mirt(data, 1, pars = svalues, verbose=FALSE)
-    expect_is(modm5, 'SingleGroupClass')
+    expect_class(modm5, 'SingleGroupClass')
     expect_warning(modm7 <- mirt(data, 1, '4PL', verbose=FALSE, parprior = list(c(3,7,11,15,19,'norm', -1.7, .1),
                                                                  c(4,8,12,16,20,'norm', 1.7, .1))),
                    "EM cycles terminated after 500 iterations.")
     expect_equal(extract.mirt(modm7, 'df'), 11)
-    expect_is(modm7, 'SingleGroupClass')
+    expect_class(modm7, 'SingleGroupClass')
     cfs <- as.numeric(do.call(c, coef(modm7)))
     expect_equal(cfs, c(5.12,8.557,0.154,0.859,5.741,3.595,0.16,0.843,10.215,10.937,0.155,0.861,1.284,0.854,0.153,0.845,4.629,8.905,0.154,0.859,0,1), tolerance = 1e-2)
     data[1,1] <- data[2,2] <- NA
     modm6 <- mirt(data, 1, verbose=FALSE)
     expect_equal(modm6@Fit$df, 21)
-    expect_is(modm6, 'SingleGroupClass')
+    expect_class(modm6, 'SingleGroupClass')
     cfs <- as.numeric(do.call(c, coef(modm6)))
     expect_equal(cfs, c(0.969, 1.851, 0, 1, 1.074, 0.808, 0, 1, 1.717, 1.811, 0, 1, 0.763, 0.486, 0, 1, 0.731, 1.852, 0, 1, 0, 1), tolerance = 1e-2)
     modideal <- mirt(data, 1, verbose=FALSE, itemtype='ideal')
@@ -93,45 +93,45 @@ test_that('dich', {
     expect_equal(as.numeric(rfit[,1]), c(NA, 0.0468866, 0.3906001, 0.2476980, 0.5195561), tolerance = 1e-2)
 
     fm1 <- fscores(modm1, verbose = FALSE, full.scores=FALSE)
-    expect_is(fm1, 'matrix')
+    expect_class(fm1, 'matrix')
     expect_true(mirt:::closeEnough(fm1[1:6,'F1'] - c(-1.8665957, -1.5266920, -1.5134024,
                                                      -1.1852276, -1.0946830, -0.7666992), -1e-2, 1e-2))
     fm2 <- fscores(modm2, method = 'MAP', verbose = FALSE, full.scores=FALSE)
-    expect_is(fm2, 'matrix')
+    expect_class(fm2, 'matrix')
     expect_true(mirt:::closeEnough(fm2[1:6,'F1'] - c(-1.8165552, -1.4946906, -1.4822982,
                                                      -1.1789899, -1.0958928, -0.7951026), -1e-2, 1e-2))
     fm3 <- fscores(modm4, method = 'ML', full.scores = TRUE, verbose = FALSE)
-    expect_is(fm3, 'matrix')
+    expect_class(fm3, 'matrix')
     expect_true(fm3[1, 'F'] == -Inf && fm3[1000, 'F'] == Inf)
     expect_true(mirt:::closeEnough(as.numeric(fm3[c(13,34,40),'F'])
                                    - c(-2.783489, -1.750890, -2.783489), -1e-2, 1e-2))
     fm3 <- fscores(modm3, method = 'ML', full.scores = TRUE, verbose = FALSE, scores.only=TRUE)
-    expect_is(fm3, 'matrix')
+    expect_class(fm3, 'matrix')
     fm4 <- fscores(modm6, method = 'ML', full.scores = TRUE, verbose = FALSE)
-    expect_is(fm4, 'matrix')
+    expect_class(fm4, 'matrix')
     fm5 <- fscores(modm6, method = 'ML', full.scores = FALSE, verbose = FALSE)
-    expect_is(fm5, 'matrix')
+    expect_class(fm5, 'matrix')
     fm6 <- fscores(modm1, method = 'EAPsum', full.scores = FALSE, verbose = FALSE)
-    expect_is(fm6, 'data.frame')
+    expect_class(fm6, 'data.frame')
     expect_true(mirt:::closeEnough(as.numeric(as.matrix(fm6)) - c(0,1,2,3,4,5,-1.86979,-1.431861,-0.9488463,-0.4131963,0.1517289,0.7271877,0.6927032,0.6838697,0.6942298,0.7210951,0.758772,0.8009335,12,40,114,205,321,308,10.08994,44.65882,109.773,207.7391,319.1854,308.5536,0.6013157,0.6971433,0.4034422,0.1900446,0.101567,0.0315184), -1e-2, 1e-2))
     expect_equal(as.numeric(attr(fm6, 'fit')['rxx_F1']), 0.4319948, tolerance = 1e-4)
 
     res1 <- residuals(modm1, verbose = FALSE)
     expect_equal(as.numeric(res1), c(NA,0.451213,0.8562096,2.577395,2.392183,-0.02124177,NA,1.053826,0.2662122,1.383089,-0.02926106,0.03246269,NA,0.1542321,0.002940504,0.05076805,-0.01631601,-0.01241902,NA,9.962506e-06,0.04890994,-0.0371899,-0.00171479,9.981236e-05,NA),
                  tolerance = 1e-2)
     res2 <- residuals(modm2, verbose = FALSE)
-    expect_is(res1, 'matrix')
-    expect_is(res2, 'matrix')
+    expect_class(res1, 'matrix')
+    expect_class(res2, 'matrix')
     IP1 <- itemplot(modm1, 1)
     IP2 <- itemplot(modm2, 1)
-    expect_is(IP1, 'trellis')
-    expect_is(IP2, 'trellis')
+    expect_class(IP1, 'trellis')
+    expect_class(IP2, 'trellis')
     TP1 <- plot(modm1)
     TP2 <- plot(modm2)
-    expect_is(TP1, 'trellis')
-    expect_is(TP2, 'trellis')
+    expect_class(TP1, 'trellis')
+    expect_class(TP2, 'trellis')
     ifit <- itemfit(modm1, c('S_X2', 'X2', 'Zh'))
-    expect_is(ifit, 'data.frame')
+    expect_class(ifit, 'data.frame')
     expect_true(mirt:::closeEnough(as.numeric(ifit$Zh) - c(1.431838, 6.354917, 5.310844, 5.804449,
                                                            0.696139), -1e-2, 1e-2))
     expect_true(mirt:::closeEnough(as.numeric(ifit$X2) - c(91.71819, 390.07985, 145.39978, 329.48529, 129.49679), -1e-2, 1e-2))
@@ -141,11 +141,11 @@ test_that('dich', {
     expect_true(mirt:::closeEnough(as.numeric(ifit$df.S_X2) - c(2,2,2,2,2), -1e-4, 1e-4))
 
     fitm1 <- M2(modm1)
-    expect_is(fitm1, 'data.frame')
+    expect_class(fitm1, 'data.frame')
     expect_true(mirt:::closeEnough(fitm1$M2 - 11.93841, -1e-2, 1e-2))
     expect_true(mirt:::closeEnough(fitm1$df.M2 - 5, -1e-4, 1e-4))
     fitm2 <- M2(modm3)
-    expect_is(fitm2, 'data.frame')
+    expect_class(fitm2, 'data.frame')
     expect_true(mirt:::closeEnough(fitm2$M2 - 5.291576, -1e-2, 1e-2))
     expect_true(mirt:::closeEnough(fitm2$df.M2 - 9, -1e-4, 1e-4))
 
@@ -160,12 +160,12 @@ test_that('dich', {
                  tolerance = 1e-2)
     expect_equal(extract.mirt(modm2, 'df'), 19)
     modm3 <- mirt(data, model, SE = TRUE, verbose=FALSE)
-    expect_is(modm3, 'SingleGroupClass')
+    expect_class(modm3, 'SingleGroupClass')
 
     fm1 <- fscores(modm1, verbose = FALSE, full.scores=FALSE)
-    expect_is(fm1, 'matrix')
+    expect_class(fm1, 'matrix')
     fm2 <- fscores(modm3, method = 'MAP', verbose = FALSE)
-    expect_is(fm2, 'matrix')
+    expect_class(fm2, 'matrix')
 
     data[1,1] <- NA
     modm1 <- mirt(data, 1, verbose=FALSE)
@@ -192,6 +192,6 @@ test_that('dich', {
                CONSTRAIN = (1-5, a1)'
     mod3 <- mirt(dat, syntax, verbose=FALSE)
     expect_equal(2*logLik(mod) - logLik(mod2) - logLik(mod3),
-                 0, 1e-2)
+                 0, tolerance=1e-2)
 })