Assumption University

630 Views

Download Presentation
## Assumption University

- - - - - - - - - - - - - - - - - - - - - - - - - - - E N D - - - - - - - - - - - - - - - - - - - - - - - - - - -

**ABAC Poll Research Center**Assumption University**ABAC POLL RESEARCH CENTER**Know your data!!! Quantitative Data Qualitative Data**ABAC POLL RESEARCH CENTER**Data Processing Data Preparation Data Analysis Data Interpretation**ABAC POLL RESEARCH CENTER**1. Close-ended Question The question that gives the choices for respondent to select. 1.1 The question that the respondents can choose only one answer, setting only one variable and the number of code will be as much as the number of choice. Ex: question concerning with sex sex ( ) 1. male ( ) 2. female variable name SEX setting:code 1 for ‘male’ and code 2 for‘female’ * The width of sex variable will be 1 column**ABAC POLL RESEARCH CENTER**1.2 The question that can be choose more than one Ex: Who is/are your favorite football players? (can choose more than one) ( ) 1. David Beckham (code 0 = dislike/ code 1 = like) ( ) 2. Michael Owen (code 0 = dislike/ code 1 = like) ( ) 3. Luis Figo (code 0 = dislike/ code 1 = like) ( ) 4. Ruud Van Nistelrooy (code 0 = dislike/ code 1 = like) ( ) 5. Raul Gonzalez (code 0 = dislike/ code 1 = like) ( ) 6. Zinedine Zidane (code 0 = dislike/ code 1 = like) ( ) 7. Michael Ballack (code 0 = dislike/ code 1 = like) ( ) 8. Ronaldo (code 0 = dislike/ code 1 = like) ( ) 9. Oliver Kahn (code 0 = dislike/ code 1 = like) ( ) 10. Ronaldinho (code 0 = dislike/ code 1 = like)**ABAC POLL RESEARCH CENTER**In this case, the variable setting will depend on the required analyzing result. If researchers want to know the number and percentage of each choice, the researchers have to set number of variables to be equal to number of choices. So, each variable will has only 2 codes; 1 selected 0 non-selected The width of each variable will be 1 column as follows: PLAYER1 : David Beckham(code 0 =non-selected / 1 = selected) PLAYER2 : Michael Owen(code 0 =non-selected / 1 = selected) PLAYER3 : Luis Figo (code 0 =non-selected / 1 = selected) Multiple Dichotomy method**ABAC POLL RESEARCH CENTER**1.3 The question that asks respondents to rank the importance A. Define each variable as a rank B. Define each variable as a choice Ex:Please rank the first three football players do you like? A. >Define variable as a rank ,total number of variables will be 3 rank1 = 1st favorite rank2 = 2nd favorite rank3 = 3rd favorite * each variable will has 10 values (number of choices)***ABAC POLL RESEARCH CENTER**B. Define each variable as a choice. Total number of variables will be 10 which equal to number of choices ( each variable will has 3 values). Ex: PLAYER1 = David Beckham (possible value is 3 ) 1 = 1st favorite 2 = 2nd favorite 3 = 3rd favorite Multiple Category method**ABAC POLL RESEARCH CENTER**2. open-ended question A>Digit data:you can setcode as value of data such as weight, height, income, and age. Ex1. How old are you?Please specify……AGE…..years old. B> Text data:you can set code for variable or keying code after you get the data. Then you have to divide data into groups and set code for each group. Ex2: what is the important problem on developing sport in Thailand? Please specify ………………..PROBLEM 1…………………**ABAC POLL RESEARCH CENTER**How to set missing data The collected questionnaires can be easily found that the respondents will leave out some questions which may intend, cannot answer, or forget to answer. In this case, the researchers have to set codeof variable that respondents leave out to be “9” or “99” or “999” depended on the width of each variable and we called it “user-missing value”**How to set variables**in questionnaires.(Distribute the questionnaire)**Exercise1**Set variables of your questionnaire and then complete the questionnaire**ABAC POLL RESEARCH CENTER**How to set up data file fromSPSS for Windows The Data Editor provides 2 views of data: Data View. Display the actual data values. VariableView. Displays variable definition information, including defined variable and value labels, data type, measurement scale and user-defined missing values. **In both views, you can add, change, and delete information contained in the data file.****ABAC POLL RESEARCH CENTER**Rows are cases. Each row represents a cases or an observation. Columns are Variables. Each column represents a variables, each item on a questionnaire is a variable. Cell Contain values. Each cell contains a single value of a variable for a case. The cell is the intersection of the case and variable.Cells contain only data values,cannot contain formulas.**ABAC POLL RESEARCH CENTER**Rows are variables. Columns are variables attributes.**ABAC POLL RESEARCH CENTER**From Variable view you can add or delete variables and modify attribute of variables, including Name :Variable name Type:Data type Width:Number of digits or characters Decimal:Number of decimal places Label:Descriptive variable Values :Value labels Missing:User-defined missing values Column:Column width Measure:Measurement scale**How to set variables**in SPSS. (Open SPSS )**Exercise2**Set variables by using SPSS.**ABAC POLL RESEARCH CENTER**Entering Data you can enter data directly in the Data Editor in the Data View.You can enter data in any oder.You can enter data by case or by variable,for selected area or for individual cells. • The active cell is highlighted. • Data values are not record until you press ‘Enter’ or select another cell. • To enter anything other than simple numeric data,you must define the variable type first. If you enter a value in an empty column,the Data Editor automatically creates a new variables and assigns a variable name as ‘var001’.**ABAC POLL RESEARCH CENTER**Editing Data in Data View with the Data Editor, you can modify data values in the Data view in many ways. You can… • Change data values. • Cut, Copy, and paste data values. • Add and delete cases. • Add and delete variables. • Change the order of variables.**Exercise3**Entering your data.**Homework**The assignment will be posted on the website by tomorrow afternoon (i.e., 3pm.)**How to Analyze your data**by using ‘SPSS’**There are 2 types of statistics**1.Descriptive Statistics 2.Inferential Statistics**ABAC POLL RESEARCH CENTER**Frequencies “The Frequencies” procedure provides statistics and graphical displays that are useful for describing many types of variables. Data: Use numeric code or short strings to code categorical variables (nominal or ordinal level measurement) To obtain frequencies and Satistics: From menus choose: Data Descriptive Statistics Frequencies…….**ABAC POLL RESEARCH CENTER**• Select one or more categorical or quantitative variables. • Optionally,you can: • Click statistics for descriptive statistics for quantitative variables. • Click Charts for bar chart, pie charts, and histograms. • Click Format for the order in which results are displayed.**ABAC POLL RESEARCH CENTER**Descriptives “The Descriptives” procedure displays univaiate summary statistics for several variables in a single table and calculates standized values (Z score).Variables can be ordered by the size of their means, alphabatically,or by order in which you select the variables (default) Data: Use numeric variables after you have screened them graphically for recoding errors, outliers. The descriptives procedure is very efficient for large file (thousands of cases). To obtain Descriptives Satistics: From menus choose: Analyze Descriptive Statistics Descriptives…….**ABAC POLL RESEARCH CENTER**• Select one or more variables. • Optionally,you can: • Click Save standized values as variables to save z score as new variables. • Click Options for optional statistics and display order.**ABAC POLL RESEARCH CENTER**Crosstabulation “The Crosstab” procedure forms two-way amd multiway tables and provides a variety of tests and measurement of the tables. Data: To define the categoriess of each table variable, use value of numeric or short string (eight or fewer charactors) var. Assumption:Some statistics and measures assume ordered categories (ordinal data) or quantitative values (interval or ratio), as discuss in the section on the statistics. To obtain Crosstabulations: From menus choose: Analyze Descriptive Statistics Crosstab…….**ABAC POLL RESEARCH CENTER**• Select one or more row variables and one or more column variables. • Optionally, you can: • Select one or more control variables. • Click Statistics for test and measures of association for two way tables or subtables. • Click Cells for observed and expected values, percentage, and residuals • Click Format for controlling the order of categories.**ABAC POLL RESEARCH CENTER**“Inference statistic” will be used to summarize data from sample group by using theory of possibility in analyzing for summarizing result to sample group. Normally, selecting of statistic depends on objective, type of variables, research form, and basic agreement of each statistic which used in such analysis. For example: inference statistic often used in general research such as population mean test, variance analysis, correlative analysis, and linear regression analysis.**ABAC POLL RESEARCH CENTER**2.1 The mean test of 1 population by using One – Sample T Test ‘The one sample T-Test’procedure tests whether the mean of a single variable differs from from a specified constant. Data:To test the values of a quantitative variable against a hypothesized test value, choose a quantitatives variable and enter a hypothesized test value. To obtain One sample T-Test: From menus choose: Analyze Compare Means One –Sample T Test…**ABAC POLL RESEARCH CENTER**Select one or more variables to be tested against the same hypothesis value. Enter a numeric test value against which each sample mean is compared.**ABAC POLL RESEARCH CENTER**2.2 The mean test of 2 independent populations by using Independent – Sample T Test Comparesthe means of one variable for two group of cases. Descriptive Statistics for each group and Levene’s test for equality of of variance are provided, as well as both equal and unequal variance t values and a 95% confidence interval for the different in means. Data:The values of the quantitative variable of interest are in a single column in the data file. The procedure uses a grouping variable with two values to separate the cases into two groups. The grouping variable can be numeric or short string.**ABAC POLL RESEARCH CENTER**To obtain Independent –Sample T Test: From menus choose: Analyze Compare Means Independent-Sample T Test… Selectone or more quantitative test variables. A separate t test is computed for each variable. Select a single grouping variable, and click Define Groups to specify two codes for the groups you want to compare.**ABAC POLL RESEARCH CENTER**2.3 The test of 2 related populations by using Paired-Sample T Test The Paired-Sample T Test procedure compares the means of two variables for a single group. It computes the differences between values of two variables for each case and test whether the average differs from 0. Data: For each paired test, specify two quantitative variables (interval or ratio scale). For a match pairs or case-control study, the response for each test subject and its matched control subject must be in the same case in data file.**ABAC POLL RESEARCH CENTER**To obtain Independent –Sample T Test: From menus choose: Analyze Compare Means Paired-Sample T Test…**ABAC POLL RESEARCH CENTER**2.4One – way analysis of variance analysis :ANOVA The One-way ANOVA procedure produces a one –way analysis of variance for a quantitative dependent variable by a single factor (independent) var. Analysis of variance is used to test the hypothesis that several means are equal. This technique is an extension of two sample t test. Data: Factor variables values should be integers, and the dependent variable should be quantitative.**ABAC POLL RESEARCH CENTER**To obtain a One-Way Analysis of Variance: From menus choose: Analyze Compare Means One-Way ANOVA… Select one or moredependent variables. Select a single independent factor variable.**ABAC POLL RESEARCH CENTER**2.5 Simple Regression Linear Regression Linear regressionestimates the coefficients of the linear equation, involving one or more independent variables, the best predict the value of the dependent variable. For example, you can try to predict a sales person’s total yearly sales (the dependent variable) from in dependent variables such as age,education, and year of experience. Data: The dependent and independent variables should be quantitative.**ABAC POLL RESEARCH CENTER**To obtain a Linear Regression Analysis : From menus choose: Analyze Regression Lenear…