update to effectsize breaks tidystats

[mattansb]

Dear @WillemSleegers,

The new version of effectsize has made a change to glass_delta() where by now a small-sample bias correction is applied.
revdepcheck shows that this change causes some tests to fail in tidystats:

      Running 'testthat.R'
     ERROR
    Running the tests in 'tests/testthat.R' failed.
    Last 13 lines of output:
          ▆
       1. └─tidystats:::expect_equal_models(model = model, expected_tidy_model = expected_statistics$glass_delta_mu) at test_effectsize.R:137:3
       2.   └─testthat::expect_equal(tidy_model, expected_tidy_model, tolerance = tolerance)
      ── Failure ('test_effectsize.R:149:3'): effectsize's Glass's delta less works ──
      `tidy_model` not equal to `expected_tidy_model`.
      Component "statistics": Component 1: Component 3: Mean relative difference: 0.06844244
      Component "statistics": Component 1: Component 7: Mean relative difference: 0.06844226
      Backtrace:
          ▆
       1. └─tidystats:::expect_equal_models(model = model, expected_tidy_model = expected_statistics$glass_delta_less) at test_effectsize.R:149:3
       2.   └─testthat::expect_equal(tidy_model, expected_tidy_model, tolerance = tolerance)
      
      [ FAIL 9 | WARN 7 | SKIP 0 | PASS 168 ]
      Error: Test failures
      Execution halted

If you wish to revert to the old behavior, you can do so with glass_delta(adjust = FALSE).

I will be pushing the update to CRAN over the next few days (easystats/effectsize#634), which will flag your package. Be advised.

Thanks!

[WillemSleegers]

Thanks for letting me know!

[mattansb]

It's on CRAN now.

[WillemSleegers]

I updated the package and it's on its way to CRAN.