higher-order constructs

Riverbank sanitation is a socioscientific issue that students in Indonesian river cities encounter first-hand, yet validated instruments for assessing how students perceive it are scarce. This study examined a 22-item instrument and how two dimensions, perceived sanitation conditions (KS; 8 items) and perceived government policy (KP; 14 items), constitute students' overall perception (PM) of sanitation on the Kuin River, Banjarmasin. Seven experts reviewed the items before administration. After screening 300 returned questionnaires and removing 49 duplicated records, 251 responses from students at three universities adjoining the Kuin riverbank were analysed. PM was specified as a reflective–formative higher-order construct and estimated with PLS-SEM (5,000 bootstrap resamples). Internal consistency was acceptable (Cronbach's α = 0.715 for KS, 0.831 for KP and 0.866 for PM; composite reliability = 0.800 and 0.865) and discriminant validity was supported (HTMT = 0.791), but average variance extracted was below 0.50 (0.33 and 0.32). Both dimensions contributed significantly to overall perception: perceived policy (β = 0.725, 95% CI [0.677, 0.774], p < .001) and perceived sanitation conditions (β = 0.374, 95% CI [0.327, 0.420], p < .001). When the two dimensions were given equal numbers of items, however, their weights converged (0.59 vs. 0.54), indicating that much of the apparent dominance of policy reflects its larger item set. Students' overall appraisal of riverbank sanitation therefore draws on both institutional and physical appraisals, with a modest edge for perceived policy. The instrument offers a starting point for diagnosing such perceptions in place-based science education, pending further validation.