Global Journal of Human Social Science, G: Linguistics and Education, Volume 23 Issue 8

Figure 1: Research model IV. M ethodology a) Research design, population, and sampling technique The study's main objective is to measure the effect of teaching quality on students' satisfaction with moderating the role of e-learning. The study used a cross-sectional survey design to achieve the study's purpose. Similarly, survey data was collected using a questionnaire adapted from previous studies. The study used final-year students from two tertiary institutions in Kano state. The population for the study is 1,620. Using final-year students became necessary as they were the only students who remained in the institution at the beginning of the industrial action embarked on by the academic staff union of universities (ASUU). The study used Krejcie & Morgan's (1970) scientific table to determine sample size. From the population of 1,620, the sample size is 310. Additionally, 10% was added to compensate for nonresponse Israel (1992) and improper filling of the instrument, raising the figure to 341. A convenience sampling technique was adopted during the data collection to enable the researchers to collect data from available and willing respondents to participate in the survey (Sekaran, 2003). Responses were recorded on a 05-point Likert scale ranging from strongly agree to strongly disagree. The study received 311 responses, but 32 surveys were eliminated because they were not completed correctly. Therefore, the study had 279 valid responses for further analysis. The response rate represented 82% of the total questionnaire administered, which was adequate for analysis. b) Measurement of variables The teaching quality (SERVQUAL) measures were adapted from (Parasuraman et al., 1988; Zeithaml et al., 1990), consisting of tangibility (TAN), reliability (REL), responsiveness (RES), assurance (ASSU), and empathy (EMP) represented by four, four, four, five, and five items. The students' satisfaction construct consisted of teaching (TCG), assessment (ASS), and generic skills and learning experience (GSLE) and was measured by six, five, and six items adapted from Fieger (2012). Finally, e-learning consists of content and educational materials (CEM), learning-teaching activities (LTA), feedback and evaluation (FE), flexibility (FXB) and infrastructure, technology and support (ITS). It was measured by five, six, five, six, and six items, each adapted from the study of (FATHI et al., 2011). V. D ata A nalysis and R esults a) Reliability and Validity This section presents the results of Cronbach's Alpha, outer loadings, composite reliability and AVE for evaluating the measurement model. The decision criterion for outer loading is 0.70, which implies that indicators with loadings below 0.70 would be deleted if the deletion could increase the reliability of the constructs (Hair et al., 2017). However, some scholars argued that loadings of 0.4 could also be considered reliable in some cases. As shown in figure 2, the AVE, composite reliability and Cronbach's Alpha values range from 0.531 to 0.969, indicating convergent validity. © 2023 Global Journals Volume XXIII Issue VIII Version I 85 ( ) Global Journal of Human Social Science - Year 2023 G Effect of Teaching Quality on Students' Satisfaction in Nigerian Tertiary Institutions: The Moderating Role of E-Learning Amid COVID-19 Recovery