The article below is about 'uji multikoliniearitas dengan spss 21' in English.

Title: Multivariate Analysis Using SPSS 21

==================================================================================

Introduction

----------

In modern statistics, researchers often encounter complex data sets with multiple continuous or categorical variables. To understand the relationships between these variables and to identify patterns within such data, multivariate analysis techniques are employed. One popular software package used for this purpose is SPSS (Statistical Package for Social Sciences) 21.

SPSS 21 provides a comprehensive suite of tools for performing various types of multivariate analyses, including exploratory factor analysis, principal component analysis, and multiple linear regression among others. This article will focus on introducing the reader to some basic concepts related to multivariate data analysis using SPSS 22 (as version 23 was not available at the time of writing) with a particular emphasis on conducting tests for multicollinearity.

Multicollinearity in Multivariate Analysis

---------------------------------------

In linear regression models, one important assumption is that all independent variables must be uncorrelated or have low levels of collinearity among them. If this condition holds true, then the estimated coefficients will capture unique effects of each predictor variable on the dependent variable(s). However, when there exists high degree of association between two or more predictors (i.e., multicollinerity), it can lead to problems such as:

1. Inflated standard errors for regression estimates

2. Non-significant p-values even in presence of real relationships

In SPSS 21, you can check for collinearity among your independent variables using the following procedures:

### Method 1: Using 'corr' Syntax Command

To perform a Pearson's correlation matrix on all numeric variables in your dataset called 'mydata', follow these steps:

1. Open IBM SPSS Statistics.

2. Click Analyze > Data Management / Transform > Correlate Cases... (for continuous data) or Variables...

3. In the 'Correlation Matrix' dialog box, select 'Pearson'.

4. Underneath the output options section, click on 'OK'.

### Method 2: Using 'corrmat()' Function in SPSS Syntax

Alternatively, you can also use the following syntax code within your SPSETUP or any other .do file:

```spss

CORRMATRIX 'correlation_matrix' = CORREL / ORDER=ASCENDING WITH variable1 variable2...

```

Interpretation of Output

----------------------

After running either method above, SPSS will produce a correlation matrix output table showing the strength and direction (positive or negative) between each pair of variables. The values range from -1 to 1 where:

* Values close to ±1 indicate strong positive/negative linear relationships.

* Zero on the diagonal indicates no relationship between a variable with itself.

For example, if we have three continuous predictor variables X, Y and Z in our dataset 'mydata', the lower-left quadrant of the output table will look something like this:

| | X | Y |

|:------:|:----------:|------------------|

| X | 1.0000 | -0.9504 |

|Y |-0.9504 | 1.0000 |

Based on these results, we can conclude that:

* There is a strong negative linear relationship between variables X and Y (correlation coefficient = -0.95).

* Variable Z does not seem to be related with either variable X or Y as indicated by the presence of zeros in its row and column.

-------

In conclusion, when working with multivariate data sets using SPSS 21, it is essential to check for multicollinearity among your independent variables before proceeding further. Failure to do so may lead to inflated standard errors and incorrect inference about relationships between predictor variables and the dependent variable(s). This article has provided an overview of two methods available in SPSPS 22 (which can also be applied with minor modifications) for detecting collinearity: using built-in syntax commands like 'corr' or utilizing functions such as corrmat(). By following these simple steps, researchers can ensure that their multivariate analyses are conducted under appropriate assumptions and yield reliable results.

Uji multikoliniearitas dengan spss 21