FM 16

• Data processing

  Involves input, throughput, and output mechanisms
• Input


  Involves the responses from the research instrument by the subjects of the study
• Throughput


  Includes statistical procedures and techniques
• Output


  The results of the study which are presented in data matrix forms
• In data processing, both quantitative and qualitative forms are involved to arrive at exact analysis and interpretation of the results
• A numerical value (quantitative) is useless without descriptive interpretation (qualitative) of the former
• Weighted arithmetic mean


  Appropriate statistical tool to determine the effectiveness of teaching
• Data processing

  Consists of categorization, coding, and tabulation of data
• Categorization of data


  • Grouping of subjects under study according to the objectives or purposes of the study
• 5 rules in categorizing research information

  • Categories are set up according to the research problem
  • The categories are exhaustive
  • The categories are mutually exclusive and independent
  • Each category (variable) is derived from one classification principle
  • Any categorization scheme must be one level of discourse
• Coding of data


  Transforming information from research instruments into coded items to facilitate tabulation
• Numerical coding is commonly used due to sufficient number coverage and is fit for computer processing
• Coding of data is more useful with research instruments having open-ended questions
• Tabulation of data


  Tallying and counting the raw data to arrive at a frequency distribution to facilitate in organizing them in a systematic order in a table or several tables
• Data matrix


  Presentation of data in tabular form, allowing for easy analysis and interpretation
• 3 types of data matrix

  • Univariate
  • Bivariate
  • Multivariate
• Dummy tables


  Helpful in preparing for the data matrix, used in planning, summarizing, organizing, and analyzing the data on how the different variables differ with each other
• Dummy tables are almost similar with real tables except that only the total number of variables, total number of cases, and percentages are presented
• Statistical treatment


  Taking raw data and turning it into something that can be interpreted and used to make decisions
• Weighted arithmetic mean is the appropriate statistical tool to use for determining the effectiveness of teaching
• Percentage is an incorrect or inappropriate statistical tool to scale options due to vague interpretation of the results
• Univariate statistical treatment


  Appropriate for both experimental and descriptive designs, using weighted arithmetic mean for scale options
• Univariate experimental research


  • Acceptability of the flavor of fish burger from offal of boneless milkfish

  
• Univariate descriptive research

  • Seriousness of job-related problems met by staff nurses in private and government hospitals

  
• Arithmetic Mean


  Used in Univariate Experimental Research
• Arithmetic Mean Calculation


  1. f

  2. x

  3. fx
• Σfx = 243, Σf = 30, Arithmetic Mean = 243/30 = 8.1
• Univariate Statistical Treatment

  • Descriptive Research
  • Weighted arithmetic mean is appropriate for scale options (i.e, 5, 4, 3, 2, and 1) and the like for univariate research problem
• Bivariate Statistical Treatment in Experimental Research

  • T-test
  • Linear Correlation
• t test

  A statistical test used to compare the means of two groups of data
• Linear Correlation


  Referred to as the measure of relationship between two random variables with values ranging from -1 and 1

  positive linear correlation when the variable on the x -axis increases as the variable on the y -axis increases
• Bivariate Statistical Treatment in Descriptive Research

  • Z-test
  • Spearman rank-coefficient of correlation or Spearman rho (rs)
• Spearman rank-coefficient of correlation or Spearman rho (rs)


  Used to determine the relationship between two variables in descriptive research
• Spearman rho Formula


  rs = 1 - (6Σ(D^2) / (N^3 - N))
• z test

  Used to determine the significant difference between two percentages of related individuals in which the data are collected through survey
• 1. test Formula


  z = (P1 - P2) / √(PQ(1/N1 + 1/N2))
• Multivariate Statistical Treatment


  • Experimental Research: F-test or ANOVA, Kruskal-Wallis One-Way Analysis of Variance, Friedman's Two-way Analysis of Variance by Ranks
  • Descriptive Research: Chi-square (X^2), Friedman's Two-way ANOVA by Ranks, Kruskal-Wallis One-Way ANOVA (H) by Ranks
• Friedman's Two-way ANOVA by Ranks Formula


  Xr^2 = (12/(NK(K+1))) * Σ(Ri)^2 - 3N(K+1)
• Chi-square (X^2) Formula


  X^2 = Σ((Observed - Expected)^2 / Expected)
• Chi-square (X^2) is used to test if variables are independent from each other in multivariate descriptive research