**7. Research findings**

#### **7.1. Homogeneity of academic achievement and intrinsic motivation**

Homogeneity analysis using T-test independent samples at 0.05 significant levels found that there were no significant difference between control and treatment groups in term of academic achievement and intrinsic motivation. **Table 3** shows pre-test mean score of academic


**Table 3.** Independent T-test pre-test mean score of academic achievement and intrinsic motivation according to groups.

achievement, t = −0.085 and df = 63.95, p > 0.05, and pre-test mean score of intrinsic motivation, t = 1.617 and df = 71, p > 0.05. The findings show that before the intervention, both the academic achievement and intrinsic motivation in the control and treatment groups were homogeneous. This allows comparison to be performed on the impact of LOC primary science module in the learning of "Plants" topic among marginalised children.

#### **7.2. Science Achievement Test (SAT)**

the format and the scope to test learners' knowledge in the topic "Plant", while IMQ was taken from Ref. [42], adapted from the Youth Children's Academic Intrinsic Motivation Inventory (Y-CAIMI) instrument in Ref. [14]. However, only two categories in IMQ were selected for this research, namely general construct and science construct. After the verification process by experts and pilot test was conducted, SAT contains 10 items, whereas IMQ contains 18 items which consists of 11 items from general constructs and 7 items from science constructs in the form of a 3-point Likert scale of "1 = Not True", "2 = Not Sure", and "3 = True". The reliability for SAT in this study showed a value of 0.711 using the Kuder Richardson approach and IMQ

After pilot test, correction and improvements was done to the module and instruments before administered them in the actual research. SAT and IMQ were administered to respondents in both groups before the T&L on plants as pre-test to determine the homogeneity level of academic achievement and intrinsic motivation between the control and treatment groups. Control group used conventional module, while treatment group used LOC module during T&L session. At the end of the T&L session, SAT and IMQ administered again to the same respondents in both groups as post-test. Both SAT and IMQ administered by the provisions of the same time taken before and after the T&L session on "Plants" topic in both control and

Quantitative data obtained through SAT and IMQ before and after the T&L session in both the control and treatment groups were analysed using descriptive statistics and inferential statistics. Independent samples T-test was conducted on the data collected during the pre-test to determine the level of homogeneity of the academic achievement and intrinsic motivation between the two groups involved. Independent samples T-test also performed on post-test to determine the effect of LOC primary science module in enhancing marginalised learners' academic achievement in science. In addition, MANOVA 2 × 2 × 2 repeated measures analysis was used to determine the effect of LOC primary science module in enhancing intrinsic motivation. Repeated measures involves two study groups (control and treatment), two time (pre-test and post-test) and two constructs of intrinsic motivation (general and science).

Homogeneity analysis using T-test independent samples at 0.05 significant levels found that there were no significant difference between control and treatment groups in term of academic achievement and intrinsic motivation. **Table 3** shows pre-test mean score of academic

**7.1. Homogeneity of academic achievement and intrinsic motivation**

showed value more than 0.70 with Cronbach alpha coefficient.

**6.4. Procedure**

66 Science Education - Research and New Technologies

treatment groups.

**7. Research findings**

**6.5. Analysis**

Before intervention, descriptive analysis found the pre-test mean score of SAT in control group, M = 46.29 (SD = 11:40), while pre-test mean score of SAT in treatment group, M = 46.58 (SD = 17.60). After intervention, descriptive analysis found the post-test mean scores of SAT in control group, M = 73.14 (SD = 21.11), while post-test mean scores of SAT in treatment group, M = 76.84 (SD = 14.91). Control group showed an increase mean score of 26.85, and treatment group showed an increase mean score of 30.26. Post-test mean scores of treatment group exceeds the control group by 3.70. **Table 4** shows the descriptive statistic of pre-test and posttest mean scores of AT according to groups.

**Table 5** shows the analysis of the independent samples T-test of post-test mean score for academic achievement according to group. Results in **Table 5** showed that there is no significant differences in the post-test mean score of SAT between the control and the treatment groups, t = −0.870 and df = 71, p > 0.05.

#### **7.3. Intrinsic motivation**

MANOVA repeated measures 2 × 2 × 2 analysis was used to determine the impact of LOC primary science module in enhancing intrinsic motivation among marginalised children


**Table 4.** Descriptive statistics pre-test and post-test mean score of achievement test according to groups.

in this research. The findings in **Table 6** showed that there is no significant main effect of group on intrinsic motivation [F (2, 70) = 0.273, p > 0.05]. Data also showed that there is no significant main effect of time on intrinsic motivation [F (2, 70) = 2.574, p > 0.05]. The effect of the interaction between time with the group is also not significant to the intrinsic motivation [F (2, 70) = 3.039, p < 0.05].

However, further analyses as shown in **Table 7** found that there is a significant main effect of the time on the general construct of intrinsic motivation [F (1, 71) = 5.054, p < 0.05]. Further descriptive analysis found that the pre-test mean score of general construct (M = 2.633, SD = 0.282) exceeds the post-test mean score of general construct (M = 2.526, SD = 0.369). This means that the level of intrinsic motivation among marginalised children generally has not been increased, but it decreased significantly across time.

The results in **Table 7** also found that there is a significant interaction effect between time and group on general construct of intrinsic motivation [F (1, 71) = 4,423, p < 0.05]. Further analysis using a paired T-test for control group general construct of intrinsic motivation was significant (t = 2.600, df = 34, p < 0.05), while the paired T-test results for the treatment group general construct of intrinsic motivation were not significant (t = 0.127, df = 37, p > 0.05). **Table 8** shows the results of paired t-test.


**Table 5.** Independent T-test post-test mean score of academic achievement according to groups.


**Table 6.** Multivariate test.


**Table 7.** Effect within subjects test.


**Table 8.** Results of paired T-test for general construct of intrinsic motivation according to time and group.
