diff --git a/DESCRIPTION b/DESCRIPTION
index 3b82c7e7dc..89af5c1bd2 100644
--- a/DESCRIPTION
+++ b/DESCRIPTION
@@ -64,7 +64,7 @@ Suggests:
     rpart,
     sf (>= 0.7-3),
     svglite (>= 1.2.0.9001),
-    testthat (>= 2.1.0),
+    testthat (>= 3.0.0),
     vdiffr (>= 1.0.0),
     xml2
 Enhances: sp
@@ -274,3 +274,4 @@ Config/Needs/website:
     tidyr,
     forcats,
     tidyverse/tidytemplate
+Config/testthat/edition: 3
diff --git a/tests/testthat/test-add.R b/tests/testthat/test-add.R
index f2dab24957..a860a55845 100644
--- a/tests/testthat/test-add.R
+++ b/tests/testthat/test-add.R
@@ -1,5 +1,3 @@
-context("Adding plot elements")
-
 test_that("mapping class is preserved when adding uneval objects", {
   p <- ggplot(mtcars) + aes(wt, mpg)
   expect_identical(class(p$mapping), "uneval")
diff --git a/tests/testthat/test-aes-calculated.r b/tests/testthat/test-aes-calculated.r
index f42584d194..8377eea5fa 100644
--- a/tests/testthat/test-aes-calculated.r
+++ b/tests/testthat/test-aes-calculated.r
@@ -1,5 +1,3 @@
-context("test-aes-calculated.r")
-
 test_that("constants aren't calculated", {
   expect_equal(is_calculated_aes(aes(1, "a", TRUE)), c(FALSE, FALSE, FALSE))
 })
diff --git a/tests/testthat/test-aes-grouping.r b/tests/testthat/test-aes-grouping.r
index f4af9b4127..2744dfd2fc 100644
--- a/tests/testthat/test-aes-grouping.r
+++ b/tests/testthat/test-aes-grouping.r
@@ -1,5 +1,3 @@
-context("Aesthetics (grouping)")
-
 df <- data_frame(
   x = 1:4,
   a = c("a", "a", "b", "b"),
diff --git a/tests/testthat/test-aes-setting.r b/tests/testthat/test-aes-setting.r
index 5dba19dfb4..baa6a12077 100644
--- a/tests/testthat/test-aes-setting.r
+++ b/tests/testthat/test-aes-setting.r
@@ -1,5 +1,3 @@
-context("Aes - setting values")
-
 test_that("aesthetic parameters match length of data", {
   df <- data_frame(x = 1:5, y = 1:5)
   p <- ggplot(df, aes(x, y))
diff --git a/tests/testthat/test-aes.r b/tests/testthat/test-aes.r
index 043ed8f641..391d1a09d3 100644
--- a/tests/testthat/test-aes.r
+++ b/tests/testthat/test-aes.r
@@ -1,5 +1,3 @@
-context("Creating aesthetic mappings")
-
 test_that("aes() captures input expressions", {
   out <- aes(mpg, wt + 1)
   expect_identical(out$x, quo(mpg))
@@ -36,7 +34,9 @@ test_that("aes_q() & aes_string() preserve explicit NULLs", {
 test_that("aes_all() converts strings into mappings", {
   expect_equal(
     aes_all(c("x", "y", "col", "pch")),
-    aes(x, y, colour = col, shape = pch)
+    aes(x, y, colour = col, shape = pch),
+    # ignore the environments of quosures
+    ignore_attr = TRUE
   )
 })
 
diff --git a/tests/testthat/test-annotate.r b/tests/testthat/test-annotate.r
index f44aee148a..6a5f36c128 100644
--- a/tests/testthat/test-annotate.r
+++ b/tests/testthat/test-annotate.r
@@ -1,5 +1,3 @@
-context("annotate")
-
 test_that("dates in segment annotation work", {
   dt <- structure(list(month = structure(c(1364774400, 1377993600),
       class = c("POSIXct", "POSIXt"), tzone = "UTC"), total = c(-10.3,
diff --git a/tests/testthat/test-build.r b/tests/testthat/test-build.r
index db89887b40..24930a0373 100644
--- a/tests/testthat/test-build.r
+++ b/tests/testthat/test-build.r
@@ -1,6 +1,4 @@
 # Test the complete path from plot specification to rendered data
-context("Plot building")
-
 df <- data_frame(x = 1:3, y = 3:1, z = letters[1:3])
 
 test_that("there is one data frame for each layer", {
@@ -19,8 +17,8 @@ test_that("position aesthetics are coerced to correct type", {
   l1 <- ggplot(df, aes(x, y)) + geom_point()
   d1 <- layer_data(l1, 1)
 
-  expect_is(d1$x, "numeric")
-  expect_is(d1$y, "numeric")
+  expect_type(d1$x, "double")
+  expect_type(d1$y, "double")
 
   l2 <- ggplot(df, aes(x, z)) + geom_point() + scale_x_discrete()
   d2 <- layer_data(l2, 1)
@@ -52,5 +50,5 @@ test_that("strings are not converted to factors", {
   p <- ggplot(df, aes(x, y)) +
     geom_text(aes(label = label), parse = TRUE)
 
-  expect_is(layer_data(p)$label, "character")
+  expect_type(layer_data(p)$label, "character")
 })
diff --git a/tests/testthat/test-coord-.r b/tests/testthat/test-coord-.r
index 4ee2f8f7dc..2d1c083a4e 100644
--- a/tests/testthat/test-coord-.r
+++ b/tests/testthat/test-coord-.r
@@ -1,5 +1,3 @@
-context("test-coord-.r")
-
 test_that("clipping is on by default", {
   p <- ggplot()
   coord <- ggplot_build(p)$layout$coord
diff --git a/tests/testthat/test-coord-cartesian.R b/tests/testthat/test-coord-cartesian.R
index c81dbff143..5b0ea913d3 100644
--- a/tests/testthat/test-coord-cartesian.R
+++ b/tests/testthat/test-coord-cartesian.R
@@ -1,5 +1,3 @@
-context("coord_cartesian")
-
 test_that("clipping can be turned off and on", {
   # clip on by default
   p <- ggplot() + coord_cartesian()
diff --git a/tests/testthat/test-coord-flip.R b/tests/testthat/test-coord-flip.R
index 0182c5d6bc..c3a24300b6 100644
--- a/tests/testthat/test-coord-flip.R
+++ b/tests/testthat/test-coord-flip.R
@@ -1,5 +1,3 @@
-context("test-coord-flip.r")
-
 test_that("secondary labels are correctly turned off", {
   # Using a visual test because the labels are only generated during rendering
   expect_doppelganger("turning off secondary title with coord_flip",
diff --git a/tests/testthat/test-coord-map.R b/tests/testthat/test-coord-map.R
index 9473820cdf..89497aba46 100644
--- a/tests/testthat/test-coord-map.R
+++ b/tests/testthat/test-coord-map.R
@@ -1,5 +1,3 @@
-context("coord_map")
-
 test_that("USA state map drawn", {
   skip_if(packageVersion("base") < "3.5.0")
   us_map <- map_data("usa")
diff --git a/tests/testthat/test-coord-polar.r b/tests/testthat/test-coord-polar.r
index 8baac118f1..c3b25ce2fa 100644
--- a/tests/testthat/test-coord-polar.r
+++ b/tests/testthat/test-coord-polar.r
@@ -1,5 +1,3 @@
-context("coord_polar")
-
 test_that("polar distance is calculated correctly", {
   dat <- data_frame(
     theta = c(0, 2*pi,   2,   6, 6, 1,    1,  0),
diff --git a/tests/testthat/test-coord-train.r b/tests/testthat/test-coord-train.r
index afca806e6c..2b712499e3 100644
--- a/tests/testthat/test-coord-train.r
+++ b/tests/testthat/test-coord-train.r
@@ -1,5 +1,3 @@
-context("coord_train")
-
 test_that("NA's don't appear in breaks", {
 
   # Returns true if any major/minor breaks have an NA
diff --git a/tests/testthat/test-coord-transform.R b/tests/testthat/test-coord-transform.R
index 1b2ecfcc11..1e2f882f87 100644
--- a/tests/testthat/test-coord-transform.R
+++ b/tests/testthat/test-coord-transform.R
@@ -1,5 +1,3 @@
-context("coord_trans")
-
 test_that("warnings are generated when cord_trans() results in new infinite values", {
   p  <- ggplot(head(diamonds, 20)) +
     geom_bar(aes(x = cut)) +
@@ -9,8 +7,12 @@ test_that("warnings are generated when cord_trans() results in new infinite valu
     geom_point() +
     coord_trans(x = "log")
 
-  expect_warning(ggplot_gtable(ggplot_build(p)), "Transformation introduced infinite values in y-axis")
-  expect_warning(ggplot_gtable(ggplot_build(p2)), "Transformation introduced infinite values in x-axis")
+  # TODO: These multiple warnings should be summarized nicely. Until this gets
+  #       fixed, this test ignores all the following errors than the first one.
+  suppressWarnings({
+    expect_warning(ggplot_gtable(ggplot_build(p)), "Transformation introduced infinite values in y-axis")
+    expect_warning(ggplot_gtable(ggplot_build(p2)), "Transformation introduced infinite values in x-axis")
+  })
 })
 
 test_that("no warnings are generated when original data has Inf values, but no new Inf values created from the transformation", {
diff --git a/tests/testthat/test-coord_sf.R b/tests/testthat/test-coord_sf.R
index fe65f72da4..3986774cea 100644
--- a/tests/testthat/test-coord_sf.R
+++ b/tests/testthat/test-coord_sf.R
@@ -1,5 +1,3 @@
-context("coord_sf")
-
 test_that("basic plot builds without error", {
   skip_if_not_installed("sf")
 
diff --git a/tests/testthat/test-data.r b/tests/testthat/test-data.r
index a363479670..57acb22544 100644
--- a/tests/testthat/test-data.r
+++ b/tests/testthat/test-data.r
@@ -1,25 +1,25 @@
-context("Data")
-
 test_that("stringsAsFactors doesn't affect results", {
-    old <- getOption("stringsAsFactors")
-    on.exit(options(stringsAsFactors = old), add = TRUE)
+  skip_if(as.integer(R.Version()$major) >= 4L, "stringsAsFactors only affects R <4.0")
+
+  old <- getOption("stringsAsFactors")
+  on.exit(options(stringsAsFactors = old), add = TRUE)
 
-    dat.character <- data_frame(x = letters[5:1], y = 1:5)
-    dat.factor <- data_frame(x = letters[5:1], y = 1:5)
+  dat.character <- data_frame(x = letters[5:1], y = 1:5)
+  dat.factor <- data_frame(x = letters[5:1], y = 1:5)
 
-    base <- ggplot(mapping = aes(x, y)) + geom_point()
-    xlabels <- function(x) x$layout$panel_params[[1]]$x$get_labels()
+  base <- ggplot(mapping = aes(x, y)) + geom_point()
+  xlabels <- function(x) x$layout$panel_params[[1]]$x$get_labels()
 
-    options(stringsAsFactors = TRUE)
-    char_true <- ggplot_build(base %+% dat.character)
-    factor_true <- ggplot_build(base %+% dat.factor)
+  options(stringsAsFactors = TRUE)
+  char_true <- ggplot_build(base %+% dat.character)
+  factor_true <- ggplot_build(base %+% dat.factor)
 
-    options(stringsAsFactors = FALSE)
-    char_false <- ggplot_build(base %+% dat.character)
-    factor_false <- ggplot_build(base %+% dat.factor)
+  options(stringsAsFactors = FALSE)
+  char_false <- ggplot_build(base %+% dat.character)
+  factor_false <- ggplot_build(base %+% dat.factor)
 
-    expect_equal(xlabels(char_true), letters[1:5])
-    expect_equal(xlabels(char_false), letters[1:5])
-    expect_equal(xlabels(factor_true), letters[1:5])
-    expect_equal(xlabels(factor_false), letters[1:5])
+  expect_equal(xlabels(char_true), letters[1:5])
+  expect_equal(xlabels(char_false), letters[1:5])
+  expect_equal(xlabels(factor_true), letters[1:5])
+  expect_equal(xlabels(factor_false), letters[1:5])
 })
diff --git a/tests/testthat/test-draw-key.R b/tests/testthat/test-draw-key.R
index 956e3755bc..d378473b20 100644
--- a/tests/testthat/test-draw-key.R
+++ b/tests/testthat/test-draw-key.R
@@ -1,6 +1,4 @@
 # Setting of legend key glyphs has to be tested visually
-context("Legend key glyphs")
-
 
 test_that("alternative key glyphs work", {
   df <- data_frame(x = 1:3, y = 3:1, z = letters[1:3])
diff --git a/tests/testthat/test-empty-data.r b/tests/testthat/test-empty-data.r
index 26c0cb8bcc..b9c72ced1b 100644
--- a/tests/testthat/test-empty-data.r
+++ b/tests/testthat/test-empty-data.r
@@ -1,5 +1,3 @@
-context('Empty data')
-
 df0 <- data_frame(mpg = numeric(0), wt = numeric(0), am = numeric(0), cyl = numeric(0))
 
 test_that("layers with empty data are silently omitted", {
diff --git a/tests/testthat/test-error.R b/tests/testthat/test-error.R
index bcf5f56c44..d845ef12e7 100644
--- a/tests/testthat/test-error.R
+++ b/tests/testthat/test-error.R
@@ -1,5 +1,3 @@
-context("error")
-
 test_that("various misuses of +.gg (#2638)", {
   expect_error(
     {
diff --git a/tests/testthat/test-facet-.r b/tests/testthat/test-facet-.r
index e54eeb7797..44a08bf1ca 100644
--- a/tests/testthat/test-facet-.r
+++ b/tests/testthat/test-facet-.r
@@ -1,5 +1,3 @@
-context("Facetting")
-
 test_that("as_facets_list() coerces formulas", {
   expect_identical(as_facets_list(~foo), list(quos(), quos(foo = foo)))
   expect_identical(as_facets_list(~foo + bar), list(quos(), quos(foo = foo, bar = bar)))
@@ -277,15 +275,16 @@ test_that("combine_vars() generates the correct combinations", {
     factor = factor(c("level1", "level2")),
     stringsAsFactors = FALSE
   )
+  attr(df_all, "out.attrs") <- NULL
 
   vars_all <- vars(letter = letter, number =  number, boolean = boolean, factor = factor)
 
-  expect_equivalent(
+  expect_equal(
     combine_vars(list(df_one), vars = vars_all),
     df_one
   )
 
-  expect_equivalent(
+  expect_equal(
     combine_vars(list(df_all), vars = vars_all),
     df_all
   )
@@ -295,21 +294,22 @@ test_that("combine_vars() generates the correct combinations", {
   # NAs are kept with with drop = TRUE
   # drop keeps all combinations of data, regardless of the combinations in which
   # they appear in the data (in addition to keeping unused factor levels)
-  expect_equivalent(
+  expect_equal(
     combine_vars(list(df_one), vars = vars_all, drop = FALSE),
-    df_all[order(df_all$letter, df_all$number, df_all$boolean, df_all$factor), ]
+    df_all[order(df_all$letter, df_all$number, df_all$boolean, df_all$factor), ],
+    ignore_attr = TRUE   # do not compare `row.names`
   )
 })
 
 test_that("drop = FALSE in combine_vars() keeps unused factor levels", {
   df <- data_frame(x = factor("a", levels = c("a", "b")))
-  expect_equivalent(
+  expect_equal(
     combine_vars(list(df), vars = vars(x = x), drop = TRUE),
-    data_frame(x = factor("a"))
+    data_frame(x = factor("a", levels = c("a", "b")))
   )
-  expect_equivalent(
+  expect_equal(
     combine_vars(list(df), vars = vars(x = x), drop = FALSE),
-    data_frame(x = factor(c("a", "b")))
+    data_frame(x = factor(c("a", "b"), levels = c("a", "b")))
   )
 })
 
diff --git a/tests/testthat/test-facet-labels.r b/tests/testthat/test-facet-labels.r
index ba87713197..6d086e0b7b 100644
--- a/tests/testthat/test-facet-labels.r
+++ b/tests/testthat/test-facet-labels.r
@@ -1,5 +1,3 @@
-context("Facet Labels")
-
 get_labels_matrix <- function(plot, ...) {
   data <- ggplot_build(plot)
   layout <- data$layout
diff --git a/tests/testthat/test-facet-layout.r b/tests/testthat/test-facet-layout.r
index c3cc9d139f..b975135557 100644
--- a/tests/testthat/test-facet-layout.r
+++ b/tests/testthat/test-facet-layout.r
@@ -1,5 +1,3 @@
-context("Facetting (layout)")
-
 a <- data_frame(a = c(1, 1, 2, 2), b = c(1, 2, 1, 1))
 b <- data_frame(a = 3)
 c <- data_frame(b = 3)
diff --git a/tests/testthat/test-facet-map.r b/tests/testthat/test-facet-map.r
index 1a0c1ba9e6..a1b10fb48b 100644
--- a/tests/testthat/test-facet-map.r
+++ b/tests/testthat/test-facet-map.r
@@ -1,5 +1,3 @@
-context("Facetting (mapping)")
-
 df <- expand.grid(a = 1:2, b = 1:2)
 df_a <- unique(df["a"])
 df_b <- unique(df["b"])
diff --git a/tests/testthat/test-facet-strips.r b/tests/testthat/test-facet-strips.r
index a9747f9f30..4536eb73d7 100644
--- a/tests/testthat/test-facet-strips.r
+++ b/tests/testthat/test-facet-strips.r
@@ -1,5 +1,3 @@
-context("Facet Strips")
-
 strip_layout <- function(p) {
   data <- ggplot_build(p)
   plot <- data$plot
diff --git a/tests/testthat/test-fortify.r b/tests/testthat/test-fortify.r
index d6510493f8..fcc0951d7d 100644
--- a/tests/testthat/test-fortify.r
+++ b/tests/testthat/test-fortify.r
@@ -1,5 +1,3 @@
-context("Fortify")
-
 test_that("spatial polygons have correct ordering", {
   skip_if_not_installed("sp")
 
@@ -31,9 +29,17 @@ test_that("spatial polygons have correct ordering", {
   polys2 <- rev(polys)
   polys2_sp <- sp::SpatialPolygons(polys2)
   fake_sp2 <- sp::SpatialPolygonsDataFrame(polys2_sp, fake_data)
-  fake_sp2_fortified <- fortify(fake_sp2)
+  expected <- fortify(fake_sp2)
+  expected <- expected[order(expected$id, expected$order), ]
+
+  actual <- fortify(fake_sp)
+
+  # the levels are different, so these columns need to be converted to character to compare
+  expected$group <- as.character(expected$group)
+  actual$group <- as.character(actual$group)
 
-  expect_equivalent(fortify(fake_sp), fake_sp2_fortified[order(fake_sp2_fortified$id, fake_sp2_fortified$order), ])
+  # Use expect_equal(ignore_attr = TRUE) to ignore rownames
+  expect_equal(actual, expected, ignore_attr = TRUE)
 })
 
 test_that("fortify.default proves a helpful error with class uneval", {
diff --git a/tests/testthat/test-function-args.r b/tests/testthat/test-function-args.r
index 080fb5b71c..5e9eb77072 100644
--- a/tests/testthat/test-function-args.r
+++ b/tests/testthat/test-function-args.r
@@ -1,5 +1,3 @@
-context("function-args")
-
 filter_args <- function(x) {
   all_names <- names(x)
   all_names <- setdiff(all_names, c("self", "data", "scales", "coordinates", "..."))
diff --git a/tests/testthat/test-geom-bar.R b/tests/testthat/test-geom-bar.R
index fa3f6bd7a3..b36a9c401e 100644
--- a/tests/testthat/test-geom-bar.R
+++ b/tests/testthat/test-geom-bar.R
@@ -1,5 +1,3 @@
-context("geom_bar")
-
 test_that("geom_bar removes bars with parts outside the plot limits", {
   dat <- data_frame(x = c("a", "b", "b", "c", "c", "c"))
 
diff --git a/tests/testthat/test-geom-boxplot.R b/tests/testthat/test-geom-boxplot.R
index d6993488bc..dd8c077877 100644
--- a/tests/testthat/test-geom-boxplot.R
+++ b/tests/testthat/test-geom-boxplot.R
@@ -1,5 +1,3 @@
-context("geom_boxplot")
-
 # thanks wch for providing the test code
 test_that("geom_boxplot range includes all outliers", {
   dat <- data_frame(x = 1, y = c(-(1:20) ^ 3, (1:20) ^ 3) )
diff --git a/tests/testthat/test-geom-col.R b/tests/testthat/test-geom-col.R
index 7d8b0548f1..d4e158ccce 100644
--- a/tests/testthat/test-geom-col.R
+++ b/tests/testthat/test-geom-col.R
@@ -1,5 +1,3 @@
-context("geom_col")
-
 test_that("geom_col removes columns with parts outside the plot limits", {
   dat <- data_frame(x = c(1, 2, 3))
 
diff --git a/tests/testthat/test-geom-dotplot.R b/tests/testthat/test-geom-dotplot.R
index e2829fec99..8049961508 100644
--- a/tests/testthat/test-geom-dotplot.R
+++ b/tests/testthat/test-geom-dotplot.R
@@ -1,5 +1,3 @@
-context("geom_dotplot")
-
 skip_on_cran() # This test suite is long-running (on cran) and is skipped
 
 set.seed(111)
diff --git a/tests/testthat/test-geom-freqpoly.R b/tests/testthat/test-geom-freqpoly.R
index 74df4d09ff..52de05f4a7 100644
--- a/tests/testthat/test-geom-freqpoly.R
+++ b/tests/testthat/test-geom-freqpoly.R
@@ -1,5 +1,3 @@
-context("geom_freqpoly")
-
 test_that("can do frequency polygon with categorical x", {
   df <- data_frame(x = rep(letters[1:3], 3:1))
 
diff --git a/tests/testthat/test-geom-hex.R b/tests/testthat/test-geom-hex.R
index d815955792..3cffeb8a66 100644
--- a/tests/testthat/test-geom-hex.R
+++ b/tests/testthat/test-geom-hex.R
@@ -1,5 +1,3 @@
-context("geom_hex")
-
 test_that("density and value summaries are available", {
   df <- data_frame(x = c(1, 1, 1, 2), y = c(1, 1, 1, 2))
   base <- ggplot(df, aes(x, y)) +
diff --git a/tests/testthat/test-geom-hline-vline-abline.R b/tests/testthat/test-geom-hline-vline-abline.R
index e57563ed1b..91b3108983 100644
--- a/tests/testthat/test-geom-hline-vline-abline.R
+++ b/tests/testthat/test-geom-hline-vline-abline.R
@@ -1,5 +1,3 @@
-context("geom-hline-vline-abline")
-
 
 # Visual tests ------------------------------------------------------------
 
diff --git a/tests/testthat/test-geom-path.R b/tests/testthat/test-geom-path.R
index 40e7b210e9..217733d9dd 100644
--- a/tests/testthat/test-geom-path.R
+++ b/tests/testthat/test-geom-path.R
@@ -1,5 +1,3 @@
-context("geom-path")
-
 test_that("keep_mid_true drops leading/trailing FALSE", {
   expect_equal(keep_mid_true(c(F, F)), c(F, F))
   expect_equal(keep_mid_true(c(F, T, F, T, F)), c(F, T, T, T, F))
diff --git a/tests/testthat/test-geom-point.R b/tests/testthat/test-geom-point.R
index b60f2b8821..bfa3300447 100644
--- a/tests/testthat/test-geom-point.R
+++ b/tests/testthat/test-geom-point.R
@@ -1,5 +1,3 @@
-context("geom-point")
-
 test_that("single strings translate to their corresponding integers", {
   expect_equal(translate_shape_string("square open"), 0)
 })
diff --git a/tests/testthat/test-geom-polygon.R b/tests/testthat/test-geom-polygon.R
index 964a8af45a..334e62043f 100644
--- a/tests/testthat/test-geom-polygon.R
+++ b/tests/testthat/test-geom-polygon.R
@@ -1,5 +1,3 @@
-context("geom-polygon")
-
 
 # Visual tests ------------------------------------------------------------
 
diff --git a/tests/testthat/test-geom-quantile.R b/tests/testthat/test-geom-quantile.R
index 61906b43a3..bdcef0ac7f 100644
--- a/tests/testthat/test-geom-quantile.R
+++ b/tests/testthat/test-geom-quantile.R
@@ -1,5 +1,3 @@
-context("geom-quantile")
-
 test_that("geom_quantile matches quantile regression", {
   skip_if_not_installed("quantreg")
 
@@ -35,25 +33,28 @@ test_that("geom_quantile matches quantile regression", {
   pred_rq_test_25 <- pred_rq[, c("x", "Q_25")]
   colnames(pred_rq_test_25) <- c("x", "y")
 
-  # Use expect_equivalent() to ignore rownames
-  expect_equivalent(
+  # Use expect_equal(ignore_attr = TRUE) to ignore rownames
+  expect_equal(
     ggplot_data[ggplot_data$quantile == 0.25, c("x", "y")],
-    pred_rq_test_25
+    pred_rq_test_25,
+    ignore_attr = TRUE
   )
 
   pred_rq_test_50 <- pred_rq[, c("x", "Q_50")]
   colnames(pred_rq_test_50) <- c("x", "y")
 
-  expect_equivalent(
+  expect_equal(
     ggplot_data[ggplot_data$quantile == 0.5, c("x", "y")],
-    pred_rq_test_50
+    pred_rq_test_50,
+    ignore_attr = TRUE
   )
 
   pred_rq_test_75 <- pred_rq[, c("x", "Q_75")]
   colnames(pred_rq_test_75) <- c("x", "y")
 
-  expect_equivalent(
+  expect_equal(
     ggplot_data[ggplot_data$quantile == 0.75, c("x", "y")],
-    pred_rq_test_75
+    pred_rq_test_75,
+    ignore_attr = TRUE
   )
 })
diff --git a/tests/testthat/test-geom-raster.R b/tests/testthat/test-geom-raster.R
index 890ef96692..cbdb9731d9 100644
--- a/tests/testthat/test-geom-raster.R
+++ b/tests/testthat/test-geom-raster.R
@@ -1,5 +1,3 @@
-context("geom-raster")
-
 
 # Visual tests ------------------------------------------------------------
 
diff --git a/tests/testthat/test-geom-ribbon.R b/tests/testthat/test-geom-ribbon.R
index 535e313651..9da04bef5a 100644
--- a/tests/testthat/test-geom-ribbon.R
+++ b/tests/testthat/test-geom-ribbon.R
@@ -1,5 +1,3 @@
-context("geom_ribbon")
-
 test_that("NAs are not dropped from the data", {
   df <- data_frame(x = 1:5, y = c(1, 1, NA, 1, 1))
 
diff --git a/tests/testthat/test-geom-rug.R b/tests/testthat/test-geom-rug.R
index 15c8d498b5..25204ef7ed 100644
--- a/tests/testthat/test-geom-rug.R
+++ b/tests/testthat/test-geom-rug.R
@@ -1,5 +1,3 @@
-context("geom_rug")
-
 n = 10
 df <- data_frame(x = 1:n, y = (1:n)^3)
 p <- ggplot(df, aes(x, y)) + geom_point() + geom_rug(sides = 'l')
diff --git a/tests/testthat/test-geom-rule.R b/tests/testthat/test-geom-rule.R
index 542957232d..e41ab41103 100644
--- a/tests/testthat/test-geom-rule.R
+++ b/tests/testthat/test-geom-rule.R
@@ -1,4 +1,3 @@
-context("geom_rule")
 # tests for geom_vline, geom_hline & geom_abline
 
 df <- data_frame(x = 1:3, y = 3:1)
diff --git a/tests/testthat/test-geom-sf.R b/tests/testthat/test-geom-sf.R
index f09e85ed3e..d3c5acc91f 100644
--- a/tests/testthat/test-geom-sf.R
+++ b/tests/testthat/test-geom-sf.R
@@ -1,5 +1,3 @@
-context("geom-sf")
-
 test_that("geom_sf() determines the legend type automatically", {
   skip_if_not_installed("sf")
   if (packageVersion("sf") < "0.5.3") skip("Need sf 0.5.3")
diff --git a/tests/testthat/test-geom-smooth.R b/tests/testthat/test-geom-smooth.R
index 12c4718014..a317d78e07 100644
--- a/tests/testthat/test-geom-smooth.R
+++ b/tests/testthat/test-geom-smooth.R
@@ -1,5 +1,3 @@
-context("geom_smooth")
-
 test_that("data is ordered by x", {
   df <- data_frame(x = c(1, 5, 2, 3, 4), y = 1:5)
 
diff --git a/tests/testthat/test-geom-text.R b/tests/testthat/test-geom-text.R
index d870a3a645..4f6573e076 100644
--- a/tests/testthat/test-geom-text.R
+++ b/tests/testthat/test-geom-text.R
@@ -1,5 +1,3 @@
-context("geom_text")
-
 # compute_just ------------------------------------------------------------
 
 test_that("vertical and horizontal positions are equivalent", {
diff --git a/tests/testthat/test-geom-tile.R b/tests/testthat/test-geom-tile.R
index be5dd39518..00261386ca 100644
--- a/tests/testthat/test-geom-tile.R
+++ b/tests/testthat/test-geom-tile.R
@@ -1,5 +1,3 @@
-context("geom_tile")
-
 test_that("accepts width and height params", {
   df <- data_frame(x = c("a", "b"), y = c("a", "b"))
 
diff --git a/tests/testthat/test-geom-violin.R b/tests/testthat/test-geom-violin.R
index 751596d3b2..d9d81b135b 100644
--- a/tests/testthat/test-geom-violin.R
+++ b/tests/testthat/test-geom-violin.R
@@ -1,5 +1,3 @@
-context("geom_violin")
-
 test_that("range is expanded", {
   df <- rbind(
     data_frame(x = "a", y = c(0, runif(10), 1)),
diff --git a/tests/testthat/test-ggproto.R b/tests/testthat/test-ggproto.R
index d6ac15b0c8..c0d4401e9e 100644
--- a/tests/testthat/test-ggproto.R
+++ b/tests/testthat/test-ggproto.R
@@ -1,5 +1,3 @@
-context("ggproto")
-
 test_that(".DollarNames retrieves inherited methods", {
   A <- ggproto("A", NULL, a = 1)
   B <- ggproto("B", A, b = 2)
diff --git a/tests/testthat/test-ggsave.R b/tests/testthat/test-ggsave.R
index 10069b4004..0d7e022d7f 100644
--- a/tests/testthat/test-ggsave.R
+++ b/tests/testthat/test-ggsave.R
@@ -1,5 +1,3 @@
-context("ggsave")
-
 test_that("ggsave creates file", {
   path <- tempfile()
   on.exit(unlink(path))
diff --git a/tests/testthat/test-grid-utils.R b/tests/testthat/test-grid-utils.R
index 2b8944b043..f9f018953e 100644
--- a/tests/testthat/test-grid-utils.R
+++ b/tests/testthat/test-grid-utils.R
@@ -1,5 +1,3 @@
-context("Grid utilites")
-
 test_that("width_cm and height_cm work with unit arithmetic", {
   x <- 2 * unit(1, "cm")
 
diff --git a/tests/testthat/test-guides.R b/tests/testthat/test-guides.R
index 541b4f3d29..c65f649794 100644
--- a/tests/testthat/test-guides.R
+++ b/tests/testthat/test-guides.R
@@ -1,5 +1,3 @@
-context("Guides")
-
 skip_on_cran() # This test suite is long-running (on cran) and is skipped
 
 test_that("colourbar trains without labels", {
@@ -60,7 +58,7 @@ test_that("axis_label_overlap_priority always returns the correct number of elem
 })
 
 test_that("axis_label_element_overrides errors when angles are outside the range [0, 90]", {
-  expect_is(axis_label_element_overrides("bottom", 0), "element")
+  expect_s3_class(axis_label_element_overrides("bottom", 0), "element")
   expect_error(axis_label_element_overrides("bottom", 91), "`angle` must")
   expect_error(axis_label_element_overrides("bottom", -91), "`angle` must")
 })
@@ -89,7 +87,12 @@ test_that("a warning is generated when more than one position guide is drawn at
       y.sec = guide_axis(position = "left")
     )
   built <- expect_silent(ggplot_build(plot))
-  expect_warning(ggplot_gtable(built), "Discarding guide")
+
+  # TODO: These multiple warnings should be summarized nicely. Until this gets
+  #       fixed, this test ignores all the following errors than the first one.
+  suppressWarnings(
+    expect_warning(ggplot_gtable(built), "Discarding guide")
+  )
 })
 
 test_that("a warning is not generated when properly changing the position of a guide_axis()", {
diff --git a/tests/testthat/test-labellers.R b/tests/testthat/test-labellers.R
index 5900e31e4e..dd6fa90ca7 100644
--- a/tests/testthat/test-labellers.R
+++ b/tests/testthat/test-labellers.R
@@ -1,5 +1,3 @@
-context("Labellers")
-
 test_that("label_bquote has access to functions in the calling environment", {
   labels <- data.frame(lab = letters[1:2])
   attr(labels, "facet") <- "wrap"
diff --git a/tests/testthat/test-labels.r b/tests/testthat/test-labels.r
index b766e0648f..0e866fa2a4 100644
--- a/tests/testthat/test-labels.r
+++ b/tests/testthat/test-labels.r
@@ -1,5 +1,3 @@
-context("Labels")
-
 test_that("setting guide labels works", {
 
     expect_identical(xlab("my label")$x, "my label")
@@ -27,16 +25,16 @@ test_that("setting guide labels works", {
     expect_identical(labs(color = "my label")$colour, "my label")
 
     # No extra elements exists
-    expect_equivalent(labs(title = "my title"), list(title = "my title"))     # formal argument
-    expect_equivalent(labs(colour = "my label"), list(colour = "my label"))   # dot
-    expect_equivalent(labs(foo = "bar"), list(foo = "bar"))                   # non-existent param
+    expect_equal(labs(title = "my title"),  list(title = "my title"),  ignore_attr = TRUE)   # formal argument
+    expect_equal(labs(colour = "my label"), list(colour = "my label"), ignore_attr = TRUE)   # dot
+    expect_equal(labs(foo = "bar"),         list(foo = "bar"),         ignore_attr = TRUE)   # non-existent param
 
     # labs() has list-splicing semantics
     params <- list(title = "my title", tag = "A)")
     expect_identical(labs(!!!params)$tag, "A)")
 
     # NULL is preserved
-    expect_equivalent(labs(title = NULL), list(title = NULL))
+    expect_equal(labs(title = NULL), list(title = NULL), ignore_attr = TRUE)
 
     # ggtitle works in the same way as labs()
     expect_identical(ggtitle("my title")$title, "my title")
@@ -44,9 +42,10 @@ test_that("setting guide labels works", {
       ggtitle("my title", subtitle = "my subtitle")$subtitle,
       "my subtitle"
     )
-    expect_equivalent(
-      ggtitle("my title", subtitle = NULL),
-      list(title = "my title", subtitle = NULL)
+    expect_equal(
+      unclass(ggtitle("my title", subtitle = NULL)),
+      list(title = "my title", subtitle = NULL),
+      ignore_attr = TRUE
     )
 })
 
diff --git a/tests/testthat/test-layer.r b/tests/testthat/test-layer.r
index e5b70eff29..ec4f60e99b 100644
--- a/tests/testthat/test-layer.r
+++ b/tests/testthat/test-layer.r
@@ -1,5 +1,3 @@
-context("Layer")
-
 # Parameters --------------------------------------------------------------
 
 test_that("aesthetics go in aes_params", {
@@ -23,7 +21,7 @@ test_that("column vectors are allowed (#2609)", {
   df <- data_frame(x = 1:10)
   df$y <- scale(1:10) # Returns a column vector
   p <- ggplot(df, aes(x, y))
-  expect_is(layer_data(p), "data.frame")
+  expect_s3_class(layer_data(p), "data.frame")
 })
 
 test_that("missing aesthetics trigger informative error", {
diff --git a/tests/testthat/test-munch.r b/tests/testthat/test-munch.r
index 8fcbb20103..5c26cab9a6 100644
--- a/tests/testthat/test-munch.r
+++ b/tests/testthat/test-munch.r
@@ -1,5 +1,3 @@
-context("Munch")
-
 test_that("interp works", {
     single_interp_test <- function(start, end, n) {
         i <- interp(start, end, n)
diff --git a/tests/testthat/test-performance.R b/tests/testthat/test-performance.R
index 5c83f2af69..3706e26cfb 100644
--- a/tests/testthat/test-performance.R
+++ b/tests/testthat/test-performance.R
@@ -1,5 +1,3 @@
-context("Performance related alternatives")
-
 # modify_list() -----------------------------------------------------------
 
 testlist <- list(
diff --git a/tests/testthat/test-plot-summary-api.R b/tests/testthat/test-plot-summary-api.R
index a8b2f7f5d9..2d45e990e7 100644
--- a/tests/testthat/test-plot-summary-api.R
+++ b/tests/testthat/test-plot-summary-api.R
@@ -1,5 +1,3 @@
-context("plot summary API")
-
 # Note: the functions tested here are used by Shiny; please do not change
 # their behavior without checking with the Shiny team first.
 
@@ -115,11 +113,11 @@ test_that("coord summary - coord_flip", {
 
 test_that("summarise_layers", {
   l <- summarise_layers(ggplot_build(p))
-  expect_equal(l$mapping[[1]], list(x = quo(displ), y = quo(hwy)))
+  expect_equal(l$mapping[[1]], list(x = quo(displ), y = quo(hwy)), ignore_attr = TRUE)
 
   p2 <- p + geom_point(aes(x = displ/2, y = hwy/2))
   l2 <- summarise_layers(ggplot_build(p2))
-  expect_equal(l2$mapping[[1]], list(x = quo(displ), y = quo(hwy)))
+  expect_equal(l2$mapping[[1]], list(x = quo(displ), y = quo(hwy)), ignore_attr = TRUE)
 
   # Here use _identical because the quosures are supposed to be local
   expect_identical(l2$mapping[[2]], list(x = quo(displ/2), y = quo(hwy/2)))
diff --git a/tests/testthat/test-position-dodge2.R b/tests/testthat/test-position-dodge2.R
index 08aa70082f..4f5fcf90dc 100644
--- a/tests/testthat/test-position-dodge2.R
+++ b/tests/testthat/test-position-dodge2.R
@@ -1,5 +1,3 @@
-context("position_dodge2")
-
 test_that("find_x_overlaps identifies overlapping groups", {
 
   df1 <- data_frame(
diff --git a/tests/testthat/test-position-nudge.R b/tests/testthat/test-position-nudge.R
index fa634aa95c..2ec0e767e2 100644
--- a/tests/testthat/test-position-nudge.R
+++ b/tests/testthat/test-position-nudge.R
@@ -1,5 +1,3 @@
-context("position_nudge")
-
 test_that("nudging works in both dimensions simultaneously", {
   # individual nudge value
   df <- data_frame(x = 1:3)
diff --git a/tests/testthat/test-position-stack.R b/tests/testthat/test-position-stack.R
index 492c850570..0243f0edd9 100644
--- a/tests/testthat/test-position-stack.R
+++ b/tests/testthat/test-position-stack.R
@@ -1,5 +1,3 @@
-context("position_stack")
-
 test_that("data keeps its order after stacking", {
   df <- data_frame(
     x = rep(c(1:10), 3),
diff --git a/tests/testthat/test-position_dodge.R b/tests/testthat/test-position_dodge.R
index ef93727763..4d540176ad 100644
--- a/tests/testthat/test-position_dodge.R
+++ b/tests/testthat/test-position_dodge.R
@@ -1,5 +1,3 @@
-context("position_dodge")
-
 test_that("can control whether to preserve total or individual width", {
   df <- data_frame(x = c("a", "b", "b"), y = c("a", "a", "b"))
 
diff --git a/tests/testthat/test-prohibited-functions.R b/tests/testthat/test-prohibited-functions.R
index 5fd3f17ff1..4303f48962 100644
--- a/tests/testthat/test-prohibited-functions.R
+++ b/tests/testthat/test-prohibited-functions.R
@@ -1,5 +1,3 @@
-context("rlang conditions")
-
 get_n_stop <- function(f) {
   d <- getParseData(parse(f, keep.source = TRUE))
   sum(d$token == "SYMBOL_FUNCTION_CALL" & d$text == "stop")
diff --git a/tests/testthat/test-qplot.r b/tests/testthat/test-qplot.r
index db56fd05d0..8e033d6397 100644
--- a/tests/testthat/test-qplot.r
+++ b/tests/testthat/test-qplot.r
@@ -1,16 +1,14 @@
-context("qplot")
-
 test_that("qplot works with variables in data frame and parent env", {
   df <- data_frame(x = 1:10, a = 1:10)
   y <- 1:10
   b <- 1:10
 
-  expect_is(qplot(x, y, data = df), "ggplot")
-  expect_is(qplot(x, y, data = df, colour = a), "ggplot")
-  expect_is(qplot(x, y, data = df, colour = b), "ggplot")
+  expect_s3_class(qplot(x, y, data = df), "ggplot")
+  expect_s3_class(qplot(x, y, data = df, colour = a), "ggplot")
+  expect_s3_class(qplot(x, y, data = df, colour = b), "ggplot")
 
   bin <- 1
-  expect_is(qplot(x, data = df, binwidth = bin), "ggplot")
+  expect_s3_class(qplot(x, data = df, binwidth = bin), "ggplot")
 })
 
 test_that("qplot works in non-standard environments", {
@@ -19,7 +17,7 @@ test_that("qplot works in non-standard environments", {
     x <- 1:10
     qplot(x, breaks = 0:`-1-`)
   })
-  expect_is(p, "ggplot")
+  expect_s3_class(p, "ggplot")
 })
 
 test_that("qplot() evaluates constants in the right place", {
diff --git a/tests/testthat/test-range.r b/tests/testthat/test-range.r
index ee33cebaf7..6e7d6b1c6c 100644
--- a/tests/testthat/test-range.r
+++ b/tests/testthat/test-range.r
@@ -1,5 +1,3 @@
-context("range")
-
 test_that("continuous ranges expand as expected", {
   r <- continuous_range()
 
diff --git a/tests/testthat/test-rbind-dfs.R b/tests/testthat/test-rbind-dfs.R
index a6c52b5e23..9184711911 100644
--- a/tests/testthat/test-rbind-dfs.R
+++ b/tests/testthat/test-rbind-dfs.R
@@ -1,5 +1,3 @@
-context("data.frame binding")
-
 test_that("rbind_dfs keep classes of columns", {
   df <- data_frame(
     integer = seq_len(10),
diff --git a/tests/testthat/test-sanitise-dim.r b/tests/testthat/test-sanitise-dim.r
index 375c434ff1..b06ca31360 100644
--- a/tests/testthat/test-sanitise-dim.r
+++ b/tests/testthat/test-sanitise-dim.r
@@ -1,5 +1,3 @@
-context("sanitise_dim")
-
 test_that("sanitise_dim returns NULL for zero-length inputs, with appropriate warnings", {
   expect_identical(sanitise_dim(NULL), NULL)
   n <- integer()
@@ -11,10 +9,16 @@ test_that("sanitise_dim returns the first element or NULL for non-positive integ
   n <- 1:2
   expect_identical(suppressWarnings(sanitise_dim(n)), 1L)
   expect_warning(sanitise_dim(n), "Only the first value of `n` will be used.")
+
   n2 <- 0:1
-  expect_identical(suppressWarnings(sanitise_dim(n2)), NULL)
-  expect_warning(sanitise_dim(n2), "Only the first value of `n2` will be used.")
-  expect_warning(sanitise_dim(n2), "`n2` is missing or less than 1 and will be treated as NULL.")
+  expect_warning(
+    expect_warning(
+      out <- sanitise_dim(n2),
+      "Only the first value of `n2` will be used."
+    ),
+    "`n2` is missing or less than 1 and will be treated as NULL."
+  )
+  expect_identical(out, NULL)
 })
 
 test_that("sanitise_dim returns a NULL for missing inputs, with appropriate warnings", {
@@ -27,8 +31,14 @@ test_that("sanitise_dim returns a positive integer or NULL for non-integer input
   n <- 1.5
   expect_identical(suppressWarnings(sanitise_dim(n)), 1L)
   expect_warning(sanitise_dim(n), "Coercing `n` to be an integer.")
+
   n2 <- 0.9999999
-  expect_identical(suppressWarnings(sanitise_dim(n2)), NULL)
-  expect_warning(sanitise_dim(n2), "Coercing `n2` to be an integer.")
-  expect_warning(sanitise_dim(n2), "`n2` is missing or less than 1 and will be treated as NULL.")
+  expect_warning(
+    expect_warning(
+      out <- sanitise_dim(n2),
+      "Coercing `n2` to be an integer."
+    ),
+    "`n2` is missing or less than 1 and will be treated as NULL."
+  )
+  expect_identical(out, NULL)
 })
diff --git a/tests/testthat/test-scale-brewer.R b/tests/testthat/test-scale-brewer.R
index 77b2c5118b..b838952fbe 100644
--- a/tests/testthat/test-scale-brewer.R
+++ b/tests/testthat/test-scale-brewer.R
@@ -1,5 +1,3 @@
-context("scale_brewer")
-
 test_that("mid-point in diverging brewer color scale", {
   d <- data_frame(x = -1:1)
 
diff --git a/tests/testthat/test-scale-colour-continuous.R b/tests/testthat/test-scale-colour-continuous.R
index 01932b715f..364bc60a91 100644
--- a/tests/testthat/test-scale-colour-continuous.R
+++ b/tests/testthat/test-scale-colour-continuous.R
@@ -1,5 +1,3 @@
-context("test-scale-colour-continuous.R")
-
 test_that("type argument is checked for proper input", {
   expect_error(
     scale_colour_continuous(type = function() "abc"),
diff --git a/tests/testthat/test-scale-date.R b/tests/testthat/test-scale-date.R
index dc08925bc7..8f0e1fa410 100644
--- a/tests/testthat/test-scale-date.R
+++ b/tests/testthat/test-scale-date.R
@@ -1,5 +1,3 @@
-context("scale_date")
-
 base_time <- function(tz = "") {
   as.POSIXct(strptime("2015-06-01", "%Y-%m-%d", tz = tz))
 }
diff --git a/tests/testthat/test-scale-discrete.R b/tests/testthat/test-scale-discrete.R
index 5d95f52dfa..d8315388ea 100644
--- a/tests/testthat/test-scale-discrete.R
+++ b/tests/testthat/test-scale-discrete.R
@@ -1,5 +1,3 @@
-context("scale_discrete")
-
 # Missing values ----------------------------------------------------------
 
 df <- tibble::tibble(
diff --git a/tests/testthat/test-scale-gradient.R b/tests/testthat/test-scale-gradient.R
index c1f6542829..fafa2226fe 100644
--- a/tests/testthat/test-scale-gradient.R
+++ b/tests/testthat/test-scale-gradient.R
@@ -1,5 +1,3 @@
-context("scale_gradient")
-
 # Limits ------------------------------------------------------------------
 
 test_that("points outside the limits are plotted as NA", {
@@ -9,5 +7,5 @@ test_that("points outside the limits are plotted as NA", {
     scale_fill_gradient2(limits = c(-1, 1), midpoint = 2, na.value = "orange")
 
   correct_fill <- c("#B26D65", "#DCB4AF", "orange")
-  expect_equivalent(layer_data(p)$fill, correct_fill)
+  expect_equal(layer_data(p)$fill, correct_fill)
 })
diff --git a/tests/testthat/test-scale-manual.r b/tests/testthat/test-scale-manual.r
index 9355558893..05da3e21d7 100644
--- a/tests/testthat/test-scale-manual.r
+++ b/tests/testthat/test-scale-manual.r
@@ -1,5 +1,3 @@
-context("scale_manual")
-
 test_that("names of values used in manual scales", {
    s1 <- scale_colour_manual(values = c("8" = "c", "4" = "a", "6" = "b"))
    s1$train(c("4", "6", "8"))
diff --git a/tests/testthat/test-scale-type.R b/tests/testthat/test-scale-type.R
index 4504df6cd0..ccefcee95b 100644
--- a/tests/testthat/test-scale-type.R
+++ b/tests/testthat/test-scale-type.R
@@ -1,5 +1,3 @@
-context("test-scale-type.R")
-
 test_that("no scale for NULL aesthetic", {
   expect_equal(find_scale("colour", NULL), NULL)
 })
diff --git a/tests/testthat/test-scale_date.R b/tests/testthat/test-scale_date.R
index 02192180b9..dafd7afffe 100644
--- a/tests/testthat/test-scale_date.R
+++ b/tests/testthat/test-scale_date.R
@@ -1,5 +1,3 @@
-context("scale_date")
-
 
 # Visual tests ------------------------------------------------------------
 
diff --git a/tests/testthat/test-scales-breaks-labels.r b/tests/testthat/test-scales-breaks-labels.r
index 3d7c2d938f..c45540fefb 100644
--- a/tests/testthat/test-scales-breaks-labels.r
+++ b/tests/testthat/test-scales-breaks-labels.r
@@ -1,5 +1,3 @@
-context("Scales: breaks and labels")
-
 test_that("labels match breaks, even when outside limits", {
   sc <- scale_y_continuous(breaks = 1:4, labels = 1:4, limits = c(1, 3))
 
@@ -99,7 +97,7 @@ test_that("discrete labels match breaks", {
   sc <- init_scale(breaks = 0:5 * 10)
   expect_equal(length(sc$get_breaks()), 5)
   expect_equal(length(sc$get_labels()), 5)
-  expect_equivalent(sc$get_labels(), sc$get_breaks())
+  expect_equal(sc$get_labels(), sc$get_breaks(), ignore_attr = TRUE)
 
   sc <- init_scale(breaks = 0:5 * 10, labels = letters[1:6])
   expect_equal(length(sc$get_breaks()), 5)
diff --git a/tests/testthat/test-scales.r b/tests/testthat/test-scales.r
index c0982662f2..308c345802 100644
--- a/tests/testthat/test-scales.r
+++ b/tests/testthat/test-scales.r
@@ -1,5 +1,3 @@
-context("Scales")
-
 test_that("building a plot does not affect its scales", {
   dat <- data_frame(x = rnorm(20), y = rnorm(20))
 
@@ -354,8 +352,8 @@ test_that("scale_apply preserves class and attributes", {
   )[[1]], `c.baz` = `c.baz`, `[.baz` = `[.baz`, .env = global_env())
 
   # Check class preservation
-  expect_is(out, "baz")
-  expect_is(out, "numeric")
+  expect_s3_class(out, "baz")
+  expect_s3_class(out, "numeric")
 
   # Check attribute preservation
   expect_identical(attr(out, "foo"), "bar")
diff --git a/tests/testthat/test-sec-axis.R b/tests/testthat/test-sec-axis.R
index 221d1468c9..ca75ca290e 100644
--- a/tests/testthat/test-sec-axis.R
+++ b/tests/testthat/test-sec-axis.R
@@ -1,5 +1,3 @@
-context("sec-axis")
-
 x <- exp(seq(log(0.001), log(1000), length.out = 100))
 foo <- data_frame(
   x = x,
diff --git a/tests/testthat/test-stat-bin.R b/tests/testthat/test-stat-bin.R
index 09f3675971..7e608e2495 100644
--- a/tests/testthat/test-stat-bin.R
+++ b/tests/testthat/test-stat-bin.R
@@ -1,5 +1,3 @@
-context("stat_bin/stat_count")
-
 test_that("stat_bin throws error when y aesthetic is present", {
   dat <- data_frame(x = c("a", "b", "c"), y = c(1, 5, 10))
 
diff --git a/tests/testthat/test-stat-bin2d.R b/tests/testthat/test-stat-bin2d.R
index 309136b8be..4f1d9d97bb 100644
--- a/tests/testthat/test-stat-bin2d.R
+++ b/tests/testthat/test-stat-bin2d.R
@@ -1,5 +1,3 @@
-context("stat_bin2d")
-
 test_that("binwidth is respected", {
   df <- data_frame(x = c(1, 1, 1, 2), y = c(1, 1, 1, 2))
   base <- ggplot(df, aes(x, y)) +
diff --git a/tests/testthat/test-stat-boxplot.R b/tests/testthat/test-stat-boxplot.R
index 7887ad7145..5e79501a95 100644
--- a/tests/testthat/test-stat-boxplot.R
+++ b/tests/testthat/test-stat-boxplot.R
@@ -1,5 +1,3 @@
-context("stat_boxplot")
-
 test_that("stat_boxplot drops missing rows with a warning", {
 
   p1 <- ggplot(PlantGrowth, aes(x = group, y = weight)) +
diff --git a/tests/testthat/test-stat-contour.R b/tests/testthat/test-stat-contour.R
index 61df774756..9843bed2b7 100644
--- a/tests/testthat/test-stat-contour.R
+++ b/tests/testthat/test-stat-contour.R
@@ -1,16 +1,21 @@
-
-context("stat-contour")
-
 test_that("a warning is issued when there is more than one z per x+y", {
   tbl <- data_frame(x = c(1, 1, 2), y = c(1, 1, 2), z = 3)
   p <- ggplot(tbl, aes(x, y, z = z)) + geom_contour()
-  expect_warning(ggplot_build(p), "Zero contours were generated")
+  # Ignore other warnings than the one stat_contour() issued
+  suppressWarnings(
+    expect_warning(ggplot_build(p), "Zero contours were generated")
+  )
 })
 
 test_that("contouring sparse data results in a warning", {
   tbl <- data_frame(x = c(1, 27, 32), y = c(1, 1, 30), z = c(1, 2, 3))
   p <- ggplot(tbl, aes(x, y, z = z)) + geom_contour()
-  expect_warning(ggplot_build(p), "Zero contours were generated")
+
+  # TODO: These multiple warnings should be summarized nicely. Until this gets
+  #       fixed, this test ignores all the following errors than the first one.
+  suppressWarnings(
+    expect_warning(ggplot_build(p), "Zero contours were generated")
+  )
 })
 
 test_that("contouring irregularly spaced data works", {
diff --git a/tests/testthat/test-stat-density.R b/tests/testthat/test-stat-density.R
index 4a26927a0c..9cc9e7a6aa 100644
--- a/tests/testthat/test-stat-density.R
+++ b/tests/testthat/test-stat-density.R
@@ -1,5 +1,3 @@
-context("stat_density") # and stat_ydensity
-
 test_that("compute_density succeeds when variance is zero", {
   dens <- compute_density(rep(0, 10), NULL, from = 0.5, to = 0.5)
   expect_equal(dens$n, rep(10, 512))
diff --git a/tests/testthat/test-stat-density2d.R b/tests/testthat/test-stat-density2d.R
index 3ac5494c90..8446feaf49 100644
--- a/tests/testthat/test-stat-density2d.R
+++ b/tests/testthat/test-stat-density2d.R
@@ -1,5 +1,3 @@
-context("stat_density_2d")
-
 test_that("uses scale limits, not data limits", {
   base <- ggplot(mtcars, aes(wt, mpg)) +
     stat_density_2d() +
diff --git a/tests/testthat/test-stat-ecdf.R b/tests/testthat/test-stat-ecdf.R
index 4e56e3625b..6c0a8d7e4d 100644
--- a/tests/testthat/test-stat-ecdf.R
+++ b/tests/testthat/test-stat-ecdf.R
@@ -1,5 +1,3 @@
-context("stat_ecdf")
-
 test_that("stat_ecdf works in both directions", {
   p <- ggplot(mpg, aes(hwy)) + stat_ecdf()
   x <- layer_data(p)
diff --git a/tests/testthat/test-stat-ellipsis.R b/tests/testthat/test-stat-ellipsis.R
index 6c1dcf8fdb..95a5df9ad7 100644
--- a/tests/testthat/test-stat-ellipsis.R
+++ b/tests/testthat/test-stat-ellipsis.R
@@ -1,5 +1,3 @@
-context("stat_ellipsis")
-
 test_that("stat_ellipsis returns correct data format", {
   n_seg <- 40
   d <- data_frame(x = c(1, 1, 4, 4, 4, 3, 3, 1), y = c(1:4, 1:4), id = rep(1:2, each = 4))
diff --git a/tests/testthat/test-stat-function.R b/tests/testthat/test-stat-function.R
index 861d88df15..fbe4f38b7e 100644
--- a/tests/testthat/test-stat-function.R
+++ b/tests/testthat/test-stat-function.R
@@ -1,5 +1,3 @@
-context("stat_function")
-
 test_that("uses scale limits, not data limits", {
   dat <- data_frame(x = c(0.1, 1:100))
   dat$y <- dexp(dat$x)
diff --git a/tests/testthat/test-stat-hex.R b/tests/testthat/test-stat-hex.R
index 1ae9a5fc63..3f9959992b 100644
--- a/tests/testthat/test-stat-hex.R
+++ b/tests/testthat/test-stat-hex.R
@@ -1,5 +1,3 @@
-context("stat_hex")
-
 test_that("can use length 1 binwidth", {
   df <- data_frame(x = c(1, 1, 2), y = c(1, 1, 2))
   p <- ggplot(df, aes(x, y)) + stat_binhex(binwidth = 1)
diff --git a/tests/testthat/test-stat-sf-coordinates.R b/tests/testthat/test-stat-sf-coordinates.R
index f328af194e..e7fa887eb5 100644
--- a/tests/testthat/test-stat-sf-coordinates.R
+++ b/tests/testthat/test-stat-sf-coordinates.R
@@ -1,5 +1,3 @@
-context("stat_sf_coordinates")
-
 comp_sf_coord <- function(df, ...) {
   plot <- ggplot(df) + stat_sf_coordinates(...)
   layer_data(plot)
diff --git a/tests/testthat/test-stat-sum.R b/tests/testthat/test-stat-sum.R
index d62584e61d..43734a0961 100644
--- a/tests/testthat/test-stat-sum.R
+++ b/tests/testthat/test-stat-sum.R
@@ -1,5 +1,3 @@
-context("stat_sum")
-
 test_that("handles grouping correctly", {
   d <- diamonds[1:1000, ]
   all_ones <- function(x) all.equal(mean(x), 1)
diff --git a/tests/testthat/test-stat-summary.R b/tests/testthat/test-stat-summary.R
index 1c4b29a166..5f8cc4b4fd 100644
--- a/tests/testthat/test-stat-summary.R
+++ b/tests/testthat/test-stat-summary.R
@@ -1,5 +1,3 @@
-context("stat_summary")
-
 test_that("stat_summary(_bin) work with lambda expressions", {
   # note: stat_summary and stat_summary_bin both use
   # make_summary_fun, so this tests both
diff --git a/tests/testthat/test-stats.r b/tests/testthat/test-stats.r
index a4be7921df..01b94a6635 100644
--- a/tests/testthat/test-stats.r
+++ b/tests/testthat/test-stats.r
@@ -1,5 +1,3 @@
-context("Stats")
-
 test_that("plot succeeds even if some computation fails", {
   df <- data_frame(x = 1:2, y = 1)
   p1 <- ggplot(df, aes(x, y)) + geom_point()
@@ -8,7 +6,12 @@ test_that("plot succeeds even if some computation fails", {
   expect_equal(length(b1$data), 1)
 
   p2 <- p1 + geom_smooth()
-  expect_warning(b2 <- ggplot_build(p2), "Computation failed")
+
+  # TODO: These multiple warnings should be summarized nicely. Until this gets
+  #       fixed, this test ignores all the following errors than the first one.
+  suppressWarnings(
+    expect_warning(b2 <- ggplot_build(p2), "Computation failed")
+  )
   expect_equal(length(b2$data), 2)
 })
 
diff --git a/tests/testthat/test-theme.r b/tests/testthat/test-theme.r
index dd06008505..1406de6ce3 100644
--- a/tests/testthat/test-theme.r
+++ b/tests/testthat/test-theme.r
@@ -1,5 +1,3 @@
-context("Themes")
-
 skip_on_cran() # This test suite is long-running (on cran) and is skipped
 
 test_that("modifying theme element properties with + operator works", {
diff --git a/tests/testthat/test-utilities.r b/tests/testthat/test-utilities.r
index e0c07e458c..7970ec5791 100644
--- a/tests/testthat/test-utilities.r
+++ b/tests/testthat/test-utilities.r
@@ -1,5 +1,3 @@
-context("Utilities")
-
 test_that("finite_cases.data.frame", {
   finite_cases <- function(x) cases(x, is_finite)
 
diff --git a/tests/testthat/test-viridis.R b/tests/testthat/test-viridis.R
index 8b4fb1e758..31dad3f3bb 100644
--- a/tests/testthat/test-viridis.R
+++ b/tests/testthat/test-viridis.R
@@ -1,5 +1,3 @@
-context("Viridis")
-
 df <- data_frame(x = 1, y = 1, z = "a", tier = factor("low", ordered = TRUE))
 
 test_that("viridis scale changes point color", {
diff --git a/tests/testthat/test-zzz.R b/tests/testthat/test-zzz.R
index 8f5ade9868..539cd88767 100644
--- a/tests/testthat/test-zzz.R
+++ b/tests/testthat/test-zzz.R
@@ -1,5 +1,3 @@
-context("zzz")
-
 test_that(".onAttach does not modify the random stream", {
   set.seed(42)
   x <- runif(5)