andkov
commented
8 years ago - Ok
- stars ar ok. If they want the specific p-vlaue, go to the dynamic tables.
- order is ok
- leave it as it is. just a simple estimate value
Open andkov opened 8 years ago
andkov
commented
8 years ago andkov
commented
8 years ago If can create EASILY a version with standardized values of the coefficients - go ahead. but not high priority.
andkov
commented
8 years ago @ampiccinin , @GracielaMuniz , @wibeasley
The rough function to pull and print a single model is ready.
> single_model_pretty <- pull_one_model(d = ds_spread_pretty,
+ study_name_ = "map",
+ subgroup_ = "female",
+ process_a_ = "grip",
+ process_b_ = "bnt",
+ model_type_ = "aehplus"
+ )
# A tibble: 41 x 4
type process label dense
<chr> <chr> <chr> <chr>
1 Fixed Effect a Intercept 43.72(1.26),*p*<.01
2 Fixed Effect a Slope -0.42(0.32),*p*=.19
3 Fixed Effect a Intercept * age -0.59(0.05),*p*<.01
4 Fixed Effect a Intercept * education 0.38(0.13),*p*<.01
5 Fixed Effect a Intercept * height 39.57(4.25),*p*<.01
6 Fixed Effect a Intercept * smoking 0.37(0.67),*p*=.58
7 Fixed Effect a Intercept * cardio 0.46(1.41),*p*=.75
8 Fixed Effect a Intercept * diabetes -0.09(1.21),*p*=.94
9 Fixed Effect a Slope * age -0.02(0.01),*p*=.06
10 Fixed Effect a Slope * education -0.07(0.03),*p*=.04
11 Fixed Effect a Slope * height -1.21(1.18),*p*=.30
12 Fixed Effect a Slope * smoking -0.01(0.17),*p*=.97
13 Fixed Effect a Slope * cardio 0.13(0.29),*p*=.64
14 Fixed Effect a Slope * diabetes -0.28(0.28),*p*=.32
15 Fixed Effect b Intercept 13.47(0.16),*p*<.01
16 Fixed Effect b Slope 0.09(0.04),*p*=.01
17 Fixed Effect b Intercept * age -0.02(0.00),*p*<.01
18 Fixed Effect b Intercept * education 0.09(0.02),*p*<.01
19 Fixed Effect b Intercept * height 2.69(0.67),*p*<.01
20 Fixed Effect b Intercept * smoking -0.01(0.08),*p*=.92
21 Fixed Effect b Intercept * cardio -0.12(0.22),*p*=.59
22 Fixed Effect b Intercept * diabetes -0.37(0.17),*p*=.03
23 Fixed Effect b Slope * age -0.00(0.00),*p*=.15
24 Fixed Effect b Slope * education -0.01(0.00),*p*=.06
25 Fixed Effect b Slope * height -0.29(0.15),*p*=.05
26 Fixed Effect b Slope * smoking -0.00(0.02),*p*=.91
27 Fixed Effect b Slope * cardio 0.03(0.05),*p*=.60
28 Fixed Effect b Slope * diabetes 0.04(0.05),*p*=.35
29 Variance a Intercept 95.59(6.02),*p*<.01
30 Variance a Slope 0.78(0.33),*p*=.02
31 Variance a Residual 19.98(1.10),*p*<.01
32 Variance b Intercept 1.21(0.19),*p*<.01
33 Variance b Slope 0.04(0.01),*p*<.01
34 Variance b Residual 0.46(0.02),*p*<.01
35 Covariance ab Intercept(a) - Intercept(b) 0.90(0.49),*p*=.07
36 Covariance ab Slope(a) - Slope(b) 0.02(0.03),*p*=.62
37 Covariance ab Intercept(a) - Slope(b) -0.02(0.13),*p*=.90
38 Covariance ab Slope(a) - Intercept(b) -0.04(0.12),*p*=.75
39 N 1,010
40 AIC 39885.414
41 BIC 40219.818See https://github.com/IALSA/IALSA-2015-Portland/commit/211d9c1d8c3ec33f086faf5306e126df856d23b4 for the script. The denses keep the actual p-values right now, although the prototype shows only stars. This was done intentionally: it is easy to switch the dense's formula, so I wanted to preserve as much detail as possible until the very end, when it's printed into a report.
The next orders of business are:
andkov
commented
8 years ago Enhanced model look-up function is now available along with the environment script ( ). The link and description of use has been added to README
> single_model <- pull_one_model(
+ d = catalog,
+ study_name_ = "map",
+ subgroup_ = "female",
+ process_a_ = "grip",
+ process_b_ = "symbol",
+ model_type_ = "aehplus",
+ pretty_ = T
+ )
# A tibble: 41 x 4
type process label dense
<chr> <chr> <chr> <chr>
1 Fixed Effect a Intercept 43.68(1.26), p<.01
2 Fixed Effect a Slope -0.41(0.32), p=.20
3 Fixed Effect a Intercept * age -0.59(0.05), p<.01
4 Fixed Effect a Intercept * education 0.38(0.13), p<.01
5 Fixed Effect a Intercept * height 39.45(4.24), p<.01
6 Fixed Effect a Intercept * smoking 0.37(0.67), p=.58
7 Fixed Effect a Intercept * cardio 0.56(1.40), p=.69
8 Fixed Effect a Intercept * diabetes -0.04(1.21), p=.97
9 Fixed Effect a Slope * age -0.02(0.01), p=.06
10 Fixed Effect a Slope * education -0.07(0.03), p=.04
11 Fixed Effect a Slope * height -1.13(1.18), p=.34
12 Fixed Effect a Slope * smoking -0.01(0.17), p=.97
13 Fixed Effect a Slope * cardio 0.13(0.29), p=.67
14 Fixed Effect a Slope * diabetes -0.29(0.28), p=.30
15 Fixed Effect b Intercept 35.43(1.05), p<.01
16 Fixed Effect b Slope 0.52(0.22), p=.02
17 Fixed Effect b Intercept * age -0.37(0.04), p<.01
18 Fixed Effect b Intercept * education 1.02(0.10), p<.01
19 Fixed Effect b Intercept * height 14.69(3.92), p<.01
20 Fixed Effect b Intercept * smoking -0.77(0.56), p=.17
21 Fixed Effect b Intercept * cardio -1.33(1.19), p=.26
22 Fixed Effect b Intercept * diabetes -3.06(1.01), p<.01
23 Fixed Effect b Slope * age -0.06(0.01), p<.01
24 Fixed Effect b Slope * education -0.05(0.02), p=.05
25 Fixed Effect b Slope * height 2.15(1.08), p=.05
26 Fixed Effect b Slope * smoking -0.17(0.14), p=.23
27 Fixed Effect b Slope * cardio 0.14(0.26), p=.59
28 Fixed Effect b Slope * diabetes -0.27(0.27), p=.32
29 Variance aa Intercept 95.67(6.03), p<.01
30 Variance aa Slope 0.79(0.33), p=.02
31 Variance a Residual 19.96(1.10), p<.01
32 Variance bb Intercept 65.55(4.59), p<.01
33 Variance bb Slope 0.98(0.31), p<.01
34 Variance b Residual 23.68(1.34), p<.01
35 Covariance ab Intercept(a) - Intercept(b) 6.24(3.52), p=.08
36 Covariance ab Slope(a) - Slope(b) -0.06(0.21), p=.79
37 Covariance ab Intercept(a) - Slope(b) 2.09(1.02), p=.04
38 Covariance ab Slope(a) - Intercept(b) 0.67(0.90), p=.46
39 <NA> <NA> N 1,010
40 <NA> <NA> AIC 53,411
41 <NA> <NA> BIC 53,745
follow up from #137
Agenda for today:
Resolutions: