Overview of methodologies. Different methodologies have been employed in order to gather data from participants from different cultures. In this section, we will first provide an overview of different methods which have been used to find answers with regards to cross-cultural differences, both in human-human, and human-robot interaction. We will then conclude with an experimental setup. A number of studies manipulated culture by using native students and exchange students in a lab experiment. ([4], [26], [34]). Already in the 80s, Xxxxxxxxxx & Xxxxxx [2] hinted that the frequent comparison of U.S. versus exchange students of a culture decreases ecological validity, because a) the students are not observed in their natural culture, b) they have been influenced by North American proxemics patterns for an undisclosed time, and c) they are not a representative sample. Xxxx et al. [39] collected data at two separate sites; thus using native students in both settings. However, this sample was also not representative because it only included students. Xxxxx et al. [40] used a method called “video-based human-robot interaction” (VHRI) in which participants viewed videos of a human interacting with a robot. Results between this video-based methodology and a lab experiment with real participants were found to be comparable. Self-reported measures, such as questionnaires, were also frequently employed. The advantage here being able to use participants from geographically distributed locations. ([3], [11], [5]). All reported studies report having the questionnaires translated and back-translated into the participants’ native language. Two experiments made use of either scaled dolls or silhouettes in order to capture people’s impression of appropriate interpersonal distance in different situations ([27], [28]). Like a lab experiment, the use of dolls does require some sort of physical location when collecting data at different sites. All these methods have advantages and disadvantages. The first method, experiments with an actual embodied robot, would be preferred for HRI since it would provide the most realistic setting. An ideal situation would be an experiment, be it a Wizard-of-Oz experiment with one type of robot, shipped all over the world to various data collection sites. This is an utopian experiment design in a world not constrained by resources like time, money and man-hours. The other methods (VHRI studies and scaled figures) could provide a solution, albeit generalizability of the re...
Overview of methodologies. At a national level, the Contractor shall conduct an analysis of the NPD for all pupils, to ascertain the relationships between scores on the Early Years Foundation Stage Profile (EYFSP), phonics check scores, phonics check ‘pass’ rates and the reading and writing levels attained at KS1, when available. These outcomes, together with information on costs, will also form the basis for VfM analysis. With a nationally representative sample of primary schools, the Contractor shall conduct a survey to establish teachers’ practices and attitudes in teaching and assessing phonics. One outcome of this shall be a typology of teachers’ engagement with the phonics agenda and the Contractor shall relate these findings to the phonics check and KS1 outcomes in the sample schools. The Contractor shall select a varied case-study sample and make a series of visits to schools, interviewing headteachers, literacy coordinators, Year 1 teachers and parents. This shall provide evidence on perceptions and attitudes which will add detail to the typology. It shall also identify what factors promote or inhibit the impact of the phonics check in encouraging the teaching of systematic synthetic phonics. A pro-forma to be completed by literacy coordinators shall examine the frequency and coverage of phonics teaching without imposing excessive burdens on the case-study schools. Over the three years of the study, the Contractors shall conduct these investigations with two cohorts of pupils. The second year of the study shall involve an intensification of data collection activities, drawing on the findings from the first year. The matrix below demonstrates how each method aligns with the research question. Research question Case studies Surveys NPD data analysis VfM analysis Year 1: the national roll-out of the check
Overview of methodologies. I developed an online survey given to studio physics students to quantify student agreement with the importance of and time spent on in-class and out-of-class activities. I made another online survey for the instructors of studio physics classes to determine what they discussed with the students about the class format, how they tried to get students to agree with the format, and the ways in which they had this discussion. Students and instructors took these surveys at the beginning and end of the semester. At the end of the semester, I collected student scores to concept tests, attitudinal surveys, and students’ expected final grades. I used hierarchical linear regressions to explore the effect of student agreement on student outcomes and the effect of instructor methods on student agreement. To triangulate (using multiple methods to support and validate my findings) the results of the surveys, I conducted interviews with instructors and students.
