*2.7. Assessments*

The main outcome measure was ROM in ankle dorsiflexion. The secondary outcome measure was COP displacements and the gait function test. All measurements were performed on the barefoot.

### *2.8. Ankle Dorsiflexion ROM*

Ankle dorsiflexion ROM was measured in the supine position and sitting position [19,39]. For the measurement, a goniometer (goniometer, jamar) was used. To measure the supine position, the participants were in the supine position on a treatment table in a knee extension posture, a sitting position. Measuring of the participants was in the sitting position and with their hip and knees' flexion in 90◦. The goniometer axis is located on the lateral malleolus and the stationary arm is located parallel to the fibular head. The movement arm was then the lateral aspect of the fifth metatarsal bone. The examiner fixed the tibial bone and pushed the foot of the participant toward the dorsiflexion. The dorsiflexion ROM was measured where the end-feel was felt, and no further movement occurred. Measurement of ankle joint ROM using a goniometer has a high reliability [40].

### *2.9. COP Displacements*

COP displacements was measured in a quiet standing position. The COP displacements evaluation is depicted in Figure 3. For the measurement, an AP1153 BioResque (RM Ingenierie, Rodez, France) was used. BioResque is a pressure force platform with 1600 sensors embedded. Participation was achieved by aligning the individuals' bare feet at the 30◦ leader line indicated above the measurement field of 400 × 400 mm, and holding the standing position for 30 s. At this time, the static sway length (cm) and static sway area (mm2) of the COP displacement value were measured. COP displacements were measured for an eyes closed and opened condition. The smaller the measured value, the better the standing balance ability.

**Figure 3.** The center of pressure (COP) displacements.

### *2.10. Gait Function*

The timed up and go test was used to assess mobility and balance. The starting position was sitting on a chair without armrests with the hip, knees, and ankle bent in a 90◦ angle. Participants had to ge<sup>t</sup> up from a chair, walk 3 m, return, and sit back in the chair. Measurements were taken after the "start" verbal cue provided by the examiner and recorded until the hips touched the chair. When measuring, the participants were barefoot. The TUG test is suitable for reliable and responsive for measuring functional mobility and dynamic balance of children with CP in GMFCS levels I–III [41]. A 10-m walk test was used to assess gait speed. In this study, a 14-m walkway was used. A stopwatch was used for the measurement, with the starting position in the standing position. When the participant started walking, the time required to walk the 10-m walking distance was measured, excluding the initial point 2 m (acceleration section) and the last point 2 m (deceleration section). The 10-m walk test provides high reliability in children with CP [42,43].

### *2.11. Statistical Analysis*

Data analysis was done using SPSS version 20 software (IBM Corp, Armonk, NY, USA). The confirmation of homogeneity and a normal distribution was verified by means of the K-S tests and independent *t*-tests and chi-square tests. The effects of intervention on dorsiflexion ROM, COP displacements, TUG, and 10-m walk test were examined with a two-way repeated-measures analysis

of variance (two-way RM ANOVA). The difference between the initial test and post-hoc test was within-group (time). The mobilization group and control group were between-group (group by time or interaction). If a significant difference appeared in the main effect or interaction, the within-group difference was measured with a paired *t*-test, whereas a between-group difference was calculated using an independent *t*-test. The alpha of statistical significance was set at 0.05.
