Statistical Analysis Project for Shahidul I.

This is not a public project, this is a private project for Shahidul Experiment Description The experiment in question, called upon indigenous males and females, belonging to different tribes around the world to participate in playing a computer game, where they had to solve a series of puzzles shown on the screen. Object selection puzzles The first puzzle category is called “object selection”. In order for each puzzle to be solved, the participant needs to read a question and select one item out of several items on the screen. For example, the user is shown a teapot, a bottle, a knife and a pen and the question is “what do you use to write on paper?”. Some object selection puzzles were related, so where they are listed in the result sheet the related ones are annotated with a letter. For example Obj_Puzzle1A is very similar to Obj_Puzzle1B, Obj_Puzzle2A is similar to Obj_Puzzle2B and so on… Sequencing puzzles The second category contains the called “sequencing puzzles”. In order for the participant to solve each puzzle, he/she will need to use the mouse to drag and drop a set of items in the correct order. For example, 3 letters are shown on the screen and the user needs to sort them into alphabetic order. The tribal breakdown The participants were 30 in total with 17 males and 13 females, split to 9 tribes from around the world. (ZULU ,APACHE,CHEROKEE,MAASSAI,MOHAWK,MAORI,SIOUX,NAVAJO,MOJAVE) Experiment Results The experiment results were exported from the software and the main items are included below. In addition to the tables, the time it took each participant to complete the try is also included for each puzzle. -- Object Selection Puzzle results ---PARTICIPANT ID The ID of each participant ---GENDER Male or female ---TRIBE The tribe the participant belongs to ---SOLVED The number of solved puzzles ---NOT SOLVED/INCORRECTLY SOLVED Then number of puzzles not solved, or incorrectly solved -- Sequencing Puzzles results ---PARTICIPANT ID The ID of each participant ---GENDER Male or female ---TRIBE The tribe the participant belongs to ---SOLVED The number of solved puzzles ---NOT SOLVED/INCORRECTLY SOLVED Then number of puzzles not solved, or incorrectly solved Your Task Your task is to do a full statistical analysis on the dataset and find hidden correlations, if any, and create a report describing the methodology you used to find said correlations. Your report should also include any sort of data you used, any software files that are required to reproduce the results, or charts and instructions on how to reproduce it (or inline comments in the code from Matlab, R, Python Scripts, SAS exports, GraphPad Prism settings,or Minitab files) It is understandable that the sample size is small so the confidence for each conclusion might be low, but this is going to be acceptable. Your general task is to derive conclusions that you believe could be statistically significant /relevant for the current experiment but some of the common-sensical questions that one would raise from running statistics in the data are mentioned below. Statistical questions 1) Search for correlations between participants tribes and puzzles by using: a. time for completion (average, total, mean et.c), b. correct or incorrect answers c. another metric? (maybe the similarity of the puzzles?). 2) Search for correlations between participants’ tribes and the two sections of the game by using: a. time for completion (average, total, mean et.c) b. correct or incorrect answers c. another metric? a) Could one correctly identify the group of participants from the available results? b) If a) is not possible, is there any common attribute(s) that characterises the participants depending on their performance on certain tasks? c) Is there any correlation between the gender of the participants/tasks/time of completion e.t.c?