Skip to main content
. Author manuscript; available in PMC: 2014 Aug 1.
Published in final edited form as: JAMA Neurol. 2013 Aug;70(8):1022–1029. doi: 10.1001/jamaneurol.2013.2934

Table 3.

Partial Least Square Regression Using Motor Tract Measurements (X) to Predict Motor Test Scores (Y)a

Behavioral Tests,
Groups
Behavioral Dyscontrol Scale 2 Purdue Pegboard Dexterity Test


Fragile X-Associated Tremor/
Ataxia Syndrome
Fragile X-Associated Tremor/
Ataxia Syndrome


Healthy Controls Negative Positive Healthy Controls Negative Positive
No. of participants 32 25 35 29 25 32

Explained variance of X, % 28.9 14.5 28.4 24.7 17.3 24.8

Explained variance of Y, % 37.0 42.3 47.0 34.9 36.6 51.5

Correlation (Y, predicted Y)

  R 0.41 0.42 0.55 0.31 0.02 0.56

  P value .02 .037b <.001 .10 .91 <.001

Paired t test (Y, predicted Y)

  df 31 24 34 28 24 31

  T −0.005 0.120 −0.094 0.006 0.176 −0.059

  95% CI −1.0 to 1.0 −1.2 to 1.3 −1.8 to 1.6 −2.4 to 2.4 −2.5 to 3.0 −2.1 to 2.0

  P value >.99 .90 .93 >.99 .86 .95
a

The results from the jackknife are shown. Bold P values are significant at a false discovery rate of 5% for Pearson correlation between Y and predicted Y (P ≤ .02).

b

Significant at a false discovery rate of 10% (P ≤ .04).