A researcher hypothesizes that witnesses who work in pairs to create a facial composite (electronic police sketch) based on a target photo of a 'suspect' will create composites that are rated (by independent judges) as looking more similar to the target photo than will facial composites created by individuals working alone. In a recent study, the 26 composites created by pairs were given an average 'similarity' rating of 5.00 with a standard deviation of 1.74 (i.e., M1 = 5.00, S1 = 1.74), while the 26 composites created by individuals were rated somewhat lower (m² = 4.23, S2 = 1.53). Are the individually-created composites rated significantly lower on similarity than those created by witnesses working in pairs? [If this is an independent samples t-test problem, use non-pooled variances] Appropriate test? H0 H1 Formula, substitution, and obtained value: Critical value(s)? Reject or fail to reject the null hypothesis? Conclusion in words?