The role of motivation and perceptions about science laboratory environment on lower secondary students’ attitude towards science

A statistical test was performed to identify the relationship between students’ attitude, motivation towards learning science and their perception of the science laboratory learning environment. A statistical data shows that attitude towards learning science was positively correlated with science laboratory learning environment (r = 0.53) and motivation towards learning science (r = 0.51). The overall model shows that learning environment and students' motivation are strongly correlated with the attitude towards learning science, with the R-value of 0.60. The high R-value indicates that the students perceived laboratory learning environment and motivation towards learning science more positively and tend to have a better attitude towards learning science. Following the correlation analysis, a multiple linear regression was performed to determine whether science laboratory learning environment and students’ motivation are the significant predictors of attitude towards learning science. The multiple linear regression indicates that both factors (learning environment and motivation) are significant predictors of the attitude (F(2, 1000) = 64.52, p < 0.05, adjusted R² = 0.36). This indicates that 36.0% of the total variance in the attitude is explained by motivation and science laboratory learning environment. This appears to be a relatively large effect as Cohen (1988) suggested that values from 0.5 to 0.8 are considered to be medium to large effect. From the result, it can be concluded that the science laboratory learning environment and students’ motivation were significant predictors of attitude towards learning science. Table 3 shows the Beta-value and the t value for the motivation and science laboratory learning environment.

Table 3. Values for SLEI and SALES in the Regression Model

Variables	Unstandardized Coefficients		Standardized Coefficients	t	Sig.
Variables	B	Std. Error	Beta	t	Sig.
SLEI	0.33	0.04	0.35	5.80	0.01
SALES	0.29	0.03	0.31	6.41	0.00

The researcher further investigated which construct in the SLEI and SALES significantly correlates with students’ attitude towards science. Results indicated that Integration, Material Environment and Students Cohesiveness from SLEI significantly correlated with students’ attitude towards learning science with R = 0.52. This reflects that secondary school students who perceived science laboratory environment to be well-equipped, promoting the integration of knowledge with scientific experiments and conducive for interactions among students tend to have a more positive attitude towards learning school science. From the five constructs in the SLEI, only three constructs (Integration, Material Environment and Students Cohesiveness) were significantly predicted the students’ science attitude (F(3, 999) = 37.56, p < 0.05, adjusted R² = 0.264). The result signified that 26.4 % of the total variance explains by these three constructs and this shows the effects to be medium (Cohen, 1988). Table 4 shows the Beta-value and the t value for the constructs in SLEI.

Table 4. Values for the Constructs in SLEI in the Regression Model

Construct	Unstandardized Coefficients		Standardized Coefficients	t	Sig.
Construct	B	Std. Error	Beta	t	Sig.
Integration	0.18	0.05	0.26	3.61	0.00
Material Environment	0.15	0.05	0.21	3.19	0.00
Students Cohesiveness	0.13	0.06	0.16	2.83	0.01

Meanwhile, for SALES, the result indicated that Task Value, Self-Efficacy and Self-Regulation have a significant correlation with students' attitude towards learning science with the R-value of 0.51. This reflects that students with high self-efficacy towards learning, high self-regulation in class and practicing high task value have a more positive attitude towards learning science. Out of four constructs, three constructs from SALES significantly predict the students’ attitude in learning science (F(3, 999) = 12.52, p < 0.05, adjusted R² = 0.256). The result signified that 25.6 % of the total variance explains by these three constructs and the effect appears to be relatively small (Cohen, 1988). Table 5 shows the Beta-value and the t value for the constructs in motivation towards learning science.

Table 5. Values for the Constructs in SALES in the Regression Model

Construct	Unstandardized Coefficients		Standardized Coefficients	t	Sig.
Construct	B	Std. Error	Beta	t	Sig.
Self-Efficacy	0.16	0.05	0.23	3.25	0.00
Self-Regulation	0.14	0.07	0.21	3.18	0.02
Task Value	0.12	0.04	0.20	3.02	0.00