Research Method: Data Treatment - Assignment Example

Research Method: Data Treatment Name: Institution: Research Method: Data Treatment DT Part A Workshop Case Study 1 Descriptive Statistics: Analyst A, Analyst B Variable Mean SE Mean StDev Variance CoefVar Q1 Median Q3 Analyst A 6.4100 0.0108 0.0216 0.0005 0.34 6.3875 6.4150 6.4275 Analyst B 6.5100 0.0387 0.0775 0.0060 1.19 6.4300 6.5300 6.5700 i) The analysis that has the most accurate value of mean is Analysis B which has a mean of 6.5100 and is closer to the certified value of 6.49. ii) The analysis that has the most precise variability or the smallest spread is Analysis A whose Standard deviation is 0.0216. The t-test results were as shown in the table below. One-Sample T: Analyst A, Analyst B Variable N Mean StDev SE Mean 95% CI Analyst A 4 6.4100 0.0216 0.0108 (6.3756, 6.4444) Analyst B 4 6.5100 0.0775 0.0387 (6.3867, 6.6333) From the t-test results, it is possible to answer the following:- ( c). There is no evidence little evidence of difference from the certified values in either case because in Analysis 1, the variation are -0.1144 and -0.0456, while in Analysis B, the variations are -0.1033 and 0.1433. (d) The results of each Analysis do not differ significantly because that of Analysis A is 0.16 while that of Analysis B is -0.2466. Hypothesis Testing The results of the hypothesis testing are as illustrated in the table below. One-Sample T: Analyst A, Analyst B Variable N Mean StDev SE Mean 6.49% CI Analyst A 4 6.4100 0.0216 0.0108 (6.4090, 6.4110) Analyst B 4 6.5100 0.0775 0.0387 (6.5066, 6.5134) For analyst A, the values of p 0.05 which shows that they are less significant. Case Study 2: One-Way Analysis of Variance The results of the ANOVA analysis were as illustrated in the figure below. One-way ANOVA: A (mL) versus B (mL) Method Null hypothesis All means are equal Alternative hypothesis At least one mean is different Significance level α = 0.05 Equal variances were assumed for the analysis. Factor Information Factor Levels Values B (mL) 3 13.79, 13.90, 13.98 Analysis of Variance Source DF Adj SS Adj MS F-Value P-Value B (mL) 2 0.004686 0.002343 * * Error 4 0.000000 0.000000 Total 6 0.004686 Model Summary S R-sq R-sq(adj) R-sq(pred) 0 100.00% 100.00% 100.00% Means B (mL) N Mean StDev 95% CI 13.79 2 14.07 0.00 (14.07, 14.07) 13.90 2 14.09 0.00 (14.09, 14.09) 13.98 3 14.03 0.00 (14.03, 14.03) Pooled StDev = 0 Case Study 3 Case Study 4 The regression outputs were as illustrated in the table below Coefficients Term Coef SE Coef T-Value P-Value VIF Constant -18.55 3.11 -5.95 0.000 X 1.509 0.124 12.21 0.000 1.00 Regression Equation Y = -18.55 + 1.509 X Case Study 5 The regression and corresponding graphs are as illustrated in the figure below. Regression Analysis: Y versus X The regression equation is Y = - 203.3 + 7.479 X S = 27.8190 R-Sq = 96.1% R-Sq(adj) = 95.7% Analysis of Variance Source DF SS MS F P Regression 1 154381 154381 199.49 0.000 Error 8 6191 774 Total 9 160572 Summary of the Data Treatment Name………………………… Student Number ……………………… Case Study 1 Basic Statistics Analyst A Analyst B Mean 6.4100 6.5100 Standard Deviation 0.0216 0.0775 Variance 0.0005 0.0060 Confidence Interval (95%) -0.1144 and -0.0456 -0.1033 and 0.1433 Which analyst is the most precise? Analysis A……………….. Reason?……Low SD…………….. Which Analyst is the most accurate?……Analysis B…………. Reason? …Mean is closer to the certified value……………… Test Null Hypothesis p Significant? A differs from standard? True 0.0108 No B differs from standard? True 0.0387 Yes A and B differ from each other? True 0.1000 No Case Study 2 H0 All means are equal ………………………… H1…… At least one mean is different ………………….. p ……=0.05………… significant? ………Yes…………………………… Does the diagram indicate anything further about the students? …………The students have similar capabilities…………………. Case Study 3: F p Significant? Temperatures 3.11 0.000 No Soils 1.509 0.000 No Interactions 1.601 0.000 No Case Study 5 Model (equation) …… Y = - 203.3 + 7.479 X …………………………………………. Lack-of-fit significant? …………Yes……………………………… Check the plot of the data and the residuals. Is there any evidence of an outlier? What further treatment of the data would you suggest? …………There is evidence of outliers………………………………………………………………………………… ……The suggestions that would be made is to improve the level of precision in collection of the data……………………………………………………………………………………… ‘Inverse Calibration problem’ Predicted x Predicted CI (m=1) Predicted CI (m=5) 7.479 X - 203.3 96.1% DT part A Option B (Biology) In biology, data treatment can be used during the establishment of a particular biological process on the selected sample of the population. This can be done by using surveys to collect data about particular sample of the population and obtain both qualitative and quantitative data. The data can then be analyzed to establish descriptive characteristics such as mean, median, mode, and standard deviation. Different software can be used to perform the analyses. These include: the use of Minitab, Statistical Software for Social Scientist (SPSS), and Microsoft Excel. The data can be interpreted based on the outcomes of the analyses. The software can also be used to perform t-tests and establish whether the hypotheses for the study are true or not. DT part A Option PE (Physicists/Engineers) Among physicists or engineers, data treatment constitutes an important process in data collection and analysis in order to understand a particular process or an activity. The process of data collection and recording can be performed using qualitative and quantitative approaches. For instance, in a case involving the analysis of the effect of temperature on a particular physical process, data treatment can involve the establishment of temperature at different stages of the process and the resulting impact on the process. This can be followed by performing descriptive statistics analysis such as establishing the mean, median, standard deviation, variance, and perform correlation analyses. DT Part A Option SQ (Survey & Questionnaire) During the study, it was initially believed that the null hypothesis is true. The null hypothesis in this case was that: Ho – There is a decrease in the level of preeclampsia condition when certain types of foods are consumed by women in the gestation period compared with others. It was assumed that the random variable was contributed by random variation only. In cases where the level of significance was 95%, the null hypothesis was rejected and if it was less than 5%, it was accepted to be true. The p-value for a null hypothesis was 0.05. Data treatment during the survey also involved the establishment of normality of data distribution such as the frequencies of occurrence of particular values of data when it was presented. The parameters that were used to describe the normal distribution included the mean (µ) and standard deviation (σ). Data treatment was also achieved by the establishment of mean. This is where the sum of all values is divided by the number of values. In the case study, it was applied by establishing the mean of participants. Data treatment was also done by establishing the media. This is the middle value when values are arranged from the smallest to the largest. Furthermore, standard deviation was used to analyze the data. This is where the level of scatter of the data is determined. This was achieved by the application of the equation: During the data analysis, a number of methods were used such as Excel worksheet and minitab. The analysis of the questionnaires was done by the use of Chi Square test of the coin tossing task to determine the level of significance of the responses. The analysis in the minitab was achieved by entering the responses in the respective columns in the sheet and the sequence of commands to follow was applied. DT Part B: Assessment and Evaluation of the project 1. Project Overview The title of the project was: Data treatment of different studies to obtain descriptive characteristics of the data. The aims of the project included the following: I. To use minitab software to determine descriptive characteristics of data such as mean, median, variance, and standard deviation. II. To use minitab software to establish the level of confidence of data collected. III. To establish whether the null hypotheses used in the study are true or false. 2. Definition of the Responses The types of responses are discrete data collected during different forms of measurements in cases 1, 2, 3, 4 and 5. The responses are quantitative because they have been presented in numeric forms that enable them to be analyzed. 3. Definition of the Factors The variables in each case are summarized in the table below. Case Study Independent Variable Dependent variable Case Study 1 Certified Value of Nickel Measured value of Nickel Case Study 2 Expected Value of titrations in millimeters Value of titrations in milliliters Case Study 3 Temperature Amount of phosphorus in soil Case Study 4 Temperature Yield of a chemical process Case Study 5 Concentration of - erylthriodine Calorimeter reading The controllable variables were: Measured value of Nickel, Value of titrations in millimeters, amount of phosphorus in soil, Yield of a Chemical process, and calorimeter reading. The uncontrollable variables were: certified value of Nickel, Expected value of titrations, temperature, and concentration of - erylthriodine. 4. Identification of the Sources of Error One of the possible sources of error is the sampling error. There is the likelihood that the sample size selected was lower than the sample size that would provide a better understanding of the entire parameters being studied. This kind of error can be minimized by increasing the sample size. Another possible sampling error during the experiment is experiment error. This is the process where the researcher does not obtain the experimental outcomes from the intended sources of data but uses the available sources of data. This sampling error can be prevented by the use of data that obtained from particular source rather than performing random selection. DT Alternative Part B 1. Project Overview The Title of the project was the ‘The use of Geographic Information System (GIS)’ application to study the state of urban development. The aim of the project was to determine the actual manner in which GIS application can be used to understand the manner in which land use occurred in the selected urban area. This was achieved by collecting secondary materials such as Google Earth maps and storing them in integrated computer software that enabled the analysis to be done. The study is being carried out because the use of GIS application has been regarded to be relevant in understanding the manner in which land use practices occur in urban areas. For instance, it enables understanding the manner in which structures such as buildings, roads, water supply systems, and commercial environments have been set up in an urban environment. In addition, the information from GIS applications can be used to implement measures for an effective management of urban development activities in a particular geographical set up. 2. Define Responses What was being measured during the project include: the quantity of different types urban land use in the location in which the study was performed such as: housing system, agricultural activities, and transport and communication system, and wildlife, and forestry activities. It also involved establishing the relationship between land use and the effectiveness of use of resources for urban development (Khan 2013). The types of responses during the project included: the names of different locations within the urban area, the names of streets, locations, industries, educational institutions, and water supply and sewerage systems. The responses were both qualitative and quantitative. Qualitative responses are those that are expressed in a descriptive manner, and represent the views, perspectives, and experiences of respondents. The main forms of qualitative data that were collected during the study include: the observations of the actual manner in which land use was done to provide a better understanding of the GIS information in the database. The recording of the actual observations of the researcher regarding the land use systems and the manner in which housing and industrial activities are organized were illustrates in qualitative form. The advantage of qualitative response is that the respondent is able to provide a response in the best way in which it is understood. The limitation of qualitative response is that it is subject to bias and the respondent may provide responses based on attitudes towards the research topic thus, affecting the validity of the research outcomes. The responses that were presented in the study were continuous in nature. In quantitative data collection, the researcher collects information that can be presented in measurable forms or quantities. In this study, the quantitative data collected included the amount of rainfall experienced in the selected urban environment, the number of housing within a particular locality, and the number of industries in the urban environment (Peck and Devore 2011). The discrete data collected was also categorical such as age groups of housings and the forms of agricultural activities practiced in the urban set up. In addition, discrete form of data collected included the actual area covered by buildings in the location and the number of industries within the focus urban area. 3. Define Factors There are a number of variables that affected the outcomes of the results. A variable is anything that changes in a measurement process. Variables can either be dependent or independent. A dependent variable is one that is changes when other variables change. In this study, the dependent variables included: planning decisions by the urban authorities, the housing activities in the location, and the existence of natural features such as vegetated areas or areas covered by physical features (Sharif et al. 2014). Independent variable is one that does not change irrespective of a change in other variables. In this study, the independent variables included human activities in the selected urban area such as educational institutions, forms of land use, and industrial activities. When the variables are ranked in terms of their impacts on the outcomes of the study, the following table provides the rank. Variable Rank Number of Houses 1 Number of highways 2 Area covered by vegetation 3 Administrative Areas 4 Learning Institutions 5 Industries 6 4. Identify Sources of Error The main source of error during the study was the random sampling error. This is the process where the sample selected does not represent the characteristics of the entire population under investigation. The sample of images collected during the GIS survey is less likely to be that of the entire urban area because some sections may be omitted while accessing them. Another sampling error is that the probability approach can be used during the selection of a sample but the actual sample may not represent the actual target geographical location being studied. For instance, if an element of sampling is not included in the sample, a sampling error occurs. This error could be overcome by the use of GIS information that covers a wider area of the geographical location being studied. Another source of sampling error during the experiment could be due to non-responsive error. This is a condition where one of the respondents who participated in a sampling process does not provide the responses despite being selected to participate. In this study, this error was possible in case the survey was conducted by means of a telephone conversation between the researcher and the respondents. Furthermore, error could result during the experiment due to measurement method used by the researcher. This is the state where the methods used to get GIS photos or images do not enable the acquisition of those explaining particular features that are of interest to the researcher. In this study, measurement error could have been caused by the acquisition of images or photographs from a different geographical location from that of the target location. 5. Presentation of the Results Different methods of presentations were used to present the outcomes of the research. An example of such an approach is the use of bar graphs. The advantage of bar graphs is that it shows different categories of data using a frequency distribution. It also enables display of numbers in proportions or categories, and clarification of trends in a better manner compared with the use of tables. The disadvantage of bar graphs is that they require additional information to provide explanations. They also have the limitation that they do not reveal major assumptions, causes, effects, and patterns existing in the data. Data presentation was also done by the use of pie charts. This is the state where the effects of various independent variables on the dependent variable were determined in graphical format. The main advantage of this method of data presentation is that it provides a display of data in a manner that is proportional to the quantity presented. It is also effective in enabling visual simplicity of the data compared with other methods of data presentation. Furthermore, pie charts can be easily understood because if their widespread in business and in the media. The limitation of pie charts is that it does not reveal the actual values and a number of them are required in order to show changes over time. Part B (Alternative) 1. Project Overview The project was titled: ‘The Application of GIS in Forest Assessment’. The aims of the study include the following: i. To establish the manner in which GIS is used to obtain information about forested areas in South Korean forests ii. To determine the manner in which South Korean Forest Service Department uses the GIS techniques to obtain photographs that enable the assessment of conditions of forests. The study was important because it enabled understanding the conditions of South Korean forests and the trends so that priorities can be made regarding the needs of rural and urban forest landscapes to enable the development of a long-term plan for the use of funds for farm management. 2. Define Responses The measured parameter during the study was the actual GIS technologies used to collect information about forests in South Korea. Another parameter that was measured is the manner in which Satellite information is used to determine the actual manner in which forest planning activities need to be done. The types of responses included: the actual areas covered by forests, how information is obtained by the use of the GIS technology, and how it is interpreted to explain the actions that need to be taken to manage the foresting activities in South Korea. 3. Define factors The variables that affected the results included: the size of areas covered by forests, the types of vegetation in different locations, and the total area where logging of trees had been done. The factors perceived to influence the outcome of the study included: the existence of planned mills, privately owned forests, zonal means in areas covered by forests, and types of vegetation found in particular forest environments. 4. Identify Sources of Errors The main sources of errors during the study could be caused by the inaccuracies of the instruments used by the GIS officials of Korea Forestry Department in the measuring of the areas covered by forests. This error can be avoided by the use of an instrument with a higher accuracy in the measurement of areas covered by forests. Another possible source of error is the interpretation of photographs from a different database rather than those that refer to Korean Forests. This error could be avoided by assessing the GIS information to validate that they refer to Korea Forests. 5. Presentation of the Results The results of the study were presented using pie charts. This was suitable for providing the comparison of the actual areas covered by different types of vegetation in South Korea. It was also of importance in ensuring the results were proportional to different forms of land use in different locations in Korea. The limitation of pie charts is that it is less suitable for providing a detailed presentation of a particular analysis. Data presentation was also done by the use of line graphs. This was mainly used in presenting the actual appearance of land such as areas with similar altitude. It is advantageous because it provides an understanding of the manner in which particular data varies over time. Its limitation is that it is less effective on presenting continuous information such as pictorial data that requires visual analysis. References Khan, R.M., 2013. Problem solving and data analysis using minitab: A clear and easy guide to six sigma methodology. John Wiley & Sons. Peck, R. and Devore, J.L., 2011. Statistics: The exploration & analysis of data. Cengage Learning. Sharif, K.M., Rahman, M.M., Azmir, J., Mohamed, A., Jahurul, M.H.A., Sahena, F. and Zaidul, I.S.M., 2014. Experimental design of supercritical fluid extraction–A review. Journal of Food Engineering, 124, pp.105-116. Thiébaux, H.J., 2013. Statistical Data Analysis for Ocean and Atmospheric Sciences: Includes a Data Disk Designed to Be Used as a Minitab File. Elsevier. Read More