SPSS statistics - get help using SPSS

Part 1: Descriptive Statistics PASW Statistics 17 (SPSS 17) ITS Training Program www.youtube.com/mycsula

Agenda Manipulating Data Selecting Cases Splitting the File Using Find and Replace Finding Data Replacing Data Reporting Copying and Pasting into Word Introduction Research Stages Opening PASW Creating a Data File Defining Variables Entering Data Running Descriptive Statistics Frequency Analysis Crosstabs

What is PASW? Predictive Analytics Software

What is Statistics? Statistics is a set of mathematical techniques used to: Summarize research data . Determine whether the data supports the researcher’s hypothesis.

Research Stages Planning and Designing Data Collecting Data Analyzing Data Reporting

Format of Questions Easy to enter Easy to construct Difficult to construct Difficult to enter Invalid responses What is your gender? a. Female b. Male What is your gender? ( _____________ ) Fixed Response Open-Ended Response e.g. PROs CONs

Running Descriptive Statistics How to analyze data. Descriptive statistics are used for summarizing frequency or measures of central tendency. Are the most commonly used statistics.

Frequency Analysis Frequency shows the number of occurrences. Also calculates measures of central tendency, such as the mean, median, mode, and others.

Research Question #1 What kind of computer do people prefer to own?

Crosstabs Crosstabs are used to examine the relationship between two variables. It shows the intersection between two variables and reveals how the two interact with each other.

Research Question #2 What color do people prefer for their computer?

Improving Your Survey What color do you like to have for your computer? 1. Beige 2. Black 3. Gray 4. White 5. Other _______

Selecting Cases Filter out and specify which variable to use for analysis with the select cases function.

Splitting the File The split file function is used to compare the responses or performance differences by groups within one variable.

Research Question #3 Is computer color preference different between genders?

Part 2: Test of Significance PASW Statistics 17 (SPSS 17) ITS Training Program www.youtube.com/mycsula

Purpose of This Workshop To show how PASW Statistics can help interpret results obtained from a sample and make inferences about the population . SAMPLE POPULATION Is it statistically significant?

Agenda Using Null Hypothesis Running Tests of Significance Correlations Paired-Samples T Test Independent-Samples T Test Running Multiple Response Sets Frequency Crosstabs Merging Data Files

A null hypothesis (H 0 ) is a statistical hypothesis that is tested for possible rejection under the assumption that it is true. The purpose of most statistical tests is to determine if the obtained results provide a reason to conclude whether or not the differences are the result of random chance. Rejection of H 0 leads to the alternative hypothesis H 1 . Null Hypothesis

Null Hypothesis The significance level ( α ) sets the standard for how extreme data must be before rejecting the H 0 . To reject H 0 , data must meet a significance level ( α ) of 0.05. α = 0.05 means data would have occurred by chance at most 5% of the time.

If p-value (sig.) ≤ α , then reject H 0 . Statistically significant If p-value (sig.) > α , then fail to reject H 0 . Statistically non-significant Hypothesis Testing Take note that the result is always stated in relation to the null hypothesis , not the alternate.

Correlations No Relationship A correlation is a statistical device that measures the nature and strength of a supposed linear association between two variables. Y X Negative Relationship Y X Y X Positive Relationship

Correlation Coefficient r = + 0.0 to 1.0 Direction Magnitude The strength of the linear relationship is determined by the distance of the correlation coefficient (r) from zero.

Research Question #1 Is there a relationship between academic performance and Internet access? H 0 = Internet access made no difference H 1 = Internet access made a different

Research Question #1 Is there a relationship between academic performance and Internet access?

T test A T test may be used to compare two group means using either one of the following: Within-participants design (a Paired-Samples T Test) Between-participants design (an Independent-Samples T Test)

Research Question #2 Is there an instructional effect taking place in the computer class? H 0 : Instruction made no difference H 1 : Instruction made a difference

Research Question #3 Is there a difference in the average number of seedlings grown in the light and those grown in the dark?

Independent-Samples T Test The first set of hypotheses is testing the variance, while the proceeding set is testing for the mean. The variances have to be equal before we can determine if the means are equal. H 0 : (µ (light) ≠ µ (dark) H 1 : (µ (light) ≠ µ (dark) H 0 : Variance (light) = variance (dark) H 1 : Variance (light) ≠ variance (dark)

Research Question #3 Is there a difference in the average number of seedlings grown in the light and those grown in the dark? H 0 : No difference whether grown in the light or dark H 1 : A difference when grown in the light versus dark

Running Multiple Response Sets Multiple response sets are used when respondents are allowed to select more than one answer in a single question. By running a frequency analysis, the result provides an overall raw frequency for each answer. Crosstabs can also be used to examine the relationship between the sets and other variables.

Merging Data Files Useful for users who store each of their topics in separate files, and eventually need or want to combine them together. This allows users to import data from one file into another. Both sets of data (from each file) must contain a common identifier for each of the cases that the user wishes to combine. An identifier identifies the correlating cases from the additional data files.

Part 3: Regression Analysis PASW Statistics 17 (SPSS 17) ITS Training Program www.youtube.com/mycsula

Purpose of This Workshop To show users how PASW Statistics can help in answering research questions or testing hypotheses by using regression. To provide users with step-by-step instructions on how to perform regression analyses with PASW Statistics.

Agenda Using Simple Regression Scatter Plot Predicting Values of Dependent Variables Predicting This Year’s Sales Using Multiple Regression Predicting Values of Dependent Variables Predicting This Year’s Sales Transforming Data Computing Using Polynomial Regression Regression Analysis Editing Charts Adding a Line Manipulating X & Y Scales Adding a Title Adding Colors Background Color

What Is Linear Regression? Linear: Straight line. Regression: Finds the model that minimizes the total variation in the data (i.e., the best fit). Linear Regression: Can be divided into two categories: Simple regression Multiple regression

What Is Polynomial Regression? Polynomial: A finite length expression constructed from variables and constants. Polynomial Regression: A special type of multiple regression used to determine the relationship between data (e .g., growth rate, progression rate).

Dependent and Independent Variables Variables can be classified into two categories: independent and dependent variables. An independent variable is a variable that influences the value of another variable. A dependent variable is a variable whose values are influenced by another variable. This is influence, not cause and effect.

Scatter Plot Before performing regression, users need to determine whether a linear relationship exists between the two variables. A scatter plot allows users to examine the linear nature of the relationship between two variables. If the relationship does not seem to be linear, then the result may be a weak regression model.

Scatter Plot Create a scatter plot to determine if a linear relationship exists between variables.

Using Simple Regression Estimates the linear relationship between one dependent ( Y ) and one independent ( X ) variable. Linear Equation: Y = a X + b a : Slope of the line b : Constant (Y-intercept, where X=0) X : Independent variable Y : Dependent variable Since we already know the values of X and Y , what we are trying to do here is to estimate a (slope) and b (Y-intercept).

Using Multiple Regression Estimates the coefficients of the linear equation, involving more than one independent variable. For example, users can predict a salesperson’s total annual sales (the dependent variable) based on independent variables, such as age, education, and years of experience.

Using Multiple Regression Linear Equation: Z = a X + b Y + c a & b : Slope coefficients c : Constant (Y-intercept) X & Y : Independent variables Z : Dependent variable

Computing Most data transformations can be done with the Compute command. Using this command, the data file can be manipulated to fit various statistical performances.

Using Polynomial Regression Variable Meaning a Constant b j The coefficient for the independent variable to the j’th power e i Random error term

Editing Charts Adding a Best Fit Line at Total

Editing Charts – Manipulating Scales

Editing Charts – Title and Gridlines

Editing Charts – Adding Colors

Part 4: Chi-Square and ANOVA PASW Statistics 17 (SPSS 17) ITS Training Program www.youtube.com/mycsula

Purpose of This Workshop To show how PASW Statistics can help answer research questions or test hypotheses by using the Chi-Square test and ANOVA. To provide step-by-step instructions on how to perform the Chi-Square test and ANOVA with PASW Statistics. To show how to import and export data using Microsoft Excel and PowerPoint. To show how to use scripting in PASW Statistics.

Agenda Using Chi-Square Test Testing for Goodness-of-Fit Using One-Way ANOVA Using Post Hoc Tests Using Two-Way ANOVA Importing/Exporting Excel Spreadsheets Using Scripting in PASW Statistics

It analyzes data in order to examine if a frequency distribution for a given variable is consistent with expectations. Chi-Square test for Goodness-of-Fit test : estimates how closely an observed distribution matches an expected distribution. Using Chi-Square Test with Fixed Expected Values

Weight Cases Before a Chi-Square test is run, weight cases should be used to identify and let PASW Statistics know what the observed values are.

Using Chi-Square Test with a Contiguous Subset

Using One-Way ANOVA ANOVA : An alysis O f Va riance. One-Way ANOVA can be thought of as a generalization of the pooled t test. Produces an analysis for a quantitative dependent variable affected by a single factor (independent variable). Instead of dealing with two populations, we have more than two populations or treatments.

Using Post Hoc Tests The null hypothesis in ANOVA is rejected when there are some differences in μ 1 , μ 2 , …, μ x . But to know where specifically these differences are, the post hoc test is used.

Using Post Hoc Tests LSD stands for List Squared Difference.

Using Two-Way ANOVA A Two-Way Analysis of Variance procedure produces an analysis for a quantitative dependent variable affected by more than one factor. It also provides information about how variables interact or combine in the effect. Advantages: More efficient Helps increase statistical power of the result

Importing/Exporting Data Data can be imported into PASW Statistics from an Excel spreadsheet. Data can be exported from PASW Statistics into an Excel spreadsheet, PowerPoint slides, etc.

Using Scripting in PASW Statistics Used to capture commands that are used repeatedly. This function simplifies working with multiple analyses on a consistent basis. Can use different data files as long as the variables in the commands always have the same name.

SPSS statistics - get help using SPSS

Recommended

More Related Content

What's hot (19)

Viewers also liked (7)

Similar to SPSS statistics - get help using SPSS (20)

Recently uploaded (20)

SPSS statistics - get help using SPSS

Editor's Notes