time analysis

Overall number of participants per local hour of the day

`geom_smooth()` using method = 'loess' and formula = 'y ~ x'

First half of the day vs second half of the day analysis

Let’s split it by the median local hour, produce the binary variable dividing the population in two halfs and check whether attention level and DG decisions differ between these two samples

Visual analysis of attention and DG_4

let’s do the same with DG decisions (DG_4)

Some analysis of naivety

`summarise()` has grouped output by 'ResponseId', 'sample'. You can override
using the `.groups` argument.

# A tibble: 2,517 × 4
# Groups:   ResponseId, sample [2,517]
   ResponseId        sample   first_half mean_naivety
   <chr>             <chr>    <fct>             <dbl>
 1 R_00QK3cjuG68NEpr Connect  Morning             0.2
 2 R_00bV8u49E8bIm1r Prolific Morning             0.8
 3 R_02PGGECILG7sgxP Besample Afternoon           0.6
 4 R_037AaDainz0D35n CR       Afternoon           0.8
 5 R_065RK8OEfxJbdG9 Toloka   Afternoon           0.6
 6 R_06WUKYA7cVDsgOR CR       Afternoon           0.8
 7 R_086w051un6Xd1hn Connect  Morning             1  
 8 R_0GUbpbjFxtwhhDP Prolific Afternoon           0.6
 9 R_0JVqLrqwBIRuxbP Toloka   Afternoon           0.4
10 R_0MJY9NSsvUNMnQZ Besample Morning             0.2
# ℹ 2,507 more rows

Warning: Removed 15 rows containing non-finite outside the scale range
(`stat_summary()`).

Warning: Removed 15 rows containing non-finite outside the scale range
(`stat_signif()`).

Warning: Removed 15 rows containing non-finite outside the scale range
(`stat_summary()`).
Removed 15 rows containing non-finite outside the scale range
(`stat_signif()`).