and emotional support (from the URES survey items). The level of significance found in this analysis promoted further investigation into the relationship of the independent variable (hall design types) on each subscale separately. Practically speaking, this finding affirmed that students do have different levels of SOC outcomes based on their experiences with select types of residence hall designs. Results of the two-way MANOVA confirmed that gender combined with residence hall design type offered statistically significant interaction effects with regard to SOC outcomes for residence hall students. Based on the exploratory nature of this analysis, Roy's Largest Root with significance at the .005 level, rather than Wilk's Lambda, was used, F(5, 344) = 3.21, p < .005; Roy's Largest Root = .06; partial eta squared = .05 (see Table 3). Interestingly, gender alone was not TABLE 2 MANOVA Sense of Community and Residence Hall Design Type EFFECT Pillai's Trace VALUE HYPOTHESIS PARTIAL ETA df ERROR df SIG. SQUARED F .083 2.48 a 12.00 688.00 .004 .041 Wilks' Lambda .919 2.48 12.00 686.00 .004 .042 Hotelling's Trace .087 2.48 .060 b a Roy's Largest Root 12.00 684.00 .004 .042 3.45 6.00 344.00 .003 .057 Exact statistic. b The statistic is an upper bound on F that yields a lower bound on the significance level. TABLE 3 MANOVA Sense of Community for Residence Hall Design Type and Gender EFFECT Pillai's Trace VALUE HYPOTHESIS PARTIAL ETA df ERROR df SIG. SQUARED F .055 1.60 a 12.00 680.00 .086 .028 Wilks' Lambda .945 1.62 12.00 678.00 .081 .028 Hotelling's Trace .058 1.64 .057 b a Roy's Largest Root 12.00 676.00 .077 .028 3.21 6.00 340.00 .004 .054 Exact statistic. b The statistic is an upper bound on F that yields a lower bound on the significance level. Volume 47, No. 2 * 2021 73