
Self-reported fitness instruments provide a scalable surveillance solution when objective testing is impractical, yet the nature of perception-accuracy discordance in adolescents is not well understood. This study examined criterion-related validity, classification accuracy, convergent construct validity against the Spanish Perceived Physical Literacy Instrument (S-PPLI), and misclassification patterning of the International Fitness Scale (IFIS) in a community sample of adolescents from the Valle de Ricote (Region of Murcia, Spain). Cross-sectional data from 1017 Spanish adolescents (471 boys, 546 girls; median age 14 years) from the Eating Healthy and Daily Life Activities (EHDLA) study were analyzed. IFIS overall correlated with the Assessing the Levels of Physical Activity and Fitness (ALPHA-Fit) composite (Spearman rho [ρ] = 0.369; partial ρ = 0.377, p < 0.001), with flexibility showing the strongest domain-specific validity (Spearman ρ = 0.635; partial ρ = 0.592, p < 0.001). Using the full five-level IFIS score, the area under the curve (AUC) for identifying the lowest fitness tertile was 0.66 (95% confidence interval [CI] 0.63-0.69). At the primary screening cut-off (IFIS ≤ 3), sensitivity was 0.68, specificity was 0.59, and negative predictive value was 0.80, while a stricter threshold (IFIS ≤ 2) increased specificity (0.94) at the cost of sensitivity (0.16). Convergent validity with S-PPLI was supported (partial ρ = 0.56, p < 0.001). Misclassification profiling identified 18.5% of participants as underestimators and 17.6% as overestimators; underestimation was associated with shorter sleep, while concordant-low membership was associated with higher body mass index (BMI), lower activity, and more sedentary time. In conclusion, IFIS showed weak-to-modest criterion-related validity and limited discriminative accuracy. Therefore, IFIS may serve as a low-burden adjunct to objective testing for population-level prioritization, but its performance relative to other brief instruments or low-burden measures, including BMI, remains untested. Discordance between perceived and objective fitness was common (36.1%) and co-occurred with lifestyle and adiposity characteristics, supporting consideration of perception-accuracy profiling in school-based and public health surveillance. Findings pertain to a single rural region and require replication in urban, regional, and socioeconomically diverse samples before broader generalization.