Bài giảng Hướng dẫn sử dụng SPSS trong nghiên cứu marketing: Regression and correlation - Ngô Thái Hưng

Chia sẻ: Dạ Du | Ngày: | Loại File: PDF | Số trang:9

Thêm vào BST

Báo xấu

11
lượt xem 1
download

Download Vui lòng tải xuống để xem tài liệu đầy đủ

Bài giảng Hướng dẫn sử dụng SPSS trong nghiên cứu marketing gồm có những nội dung chính sau: Regression and correlation, scatter diagram, correlation, regression, multiple regression. Mời các bạn cùng tham khảo để biết thêm nội dung chi tiết!

Chủ đề:

Bình luận(0) Đăng nhập để gửi bình luận!

Lưu

Nội dung Text: Bài giảng Hướng dẫn sử dụng SPSS trong nghiên cứu marketing: Regression and correlation - Ngô Thái Hưng

Đ I H C TÀI CHÍNH - MARKETING Chuyên đ : HƯ NG D N S D NG SPSS TRONG NGHIÊN C U MARKETING Ngô Thái Hưng
REGRESSION AND CORRELATION
INTRODUCTION In statistics when faced with data we attempt to summarise it and then look for the patterns. Regression is about patterns, the possible relationship between two sets of data, bivariate data
INTRODUCTION Are two variables related? What sort of relationship is there? Can we describe the strength of the relationship? Can we predict one variable from the other?
SCATTER DIAGRAM The first thing to do is to plot the data, with the independent (x) variable on the horizontal axis and the dependent (y) variable vertically. Graphs > Chart > Scatter/Dot
CORRELATION Corelation quantifies the strength of the linear relationship between the two variables and olso indicates the direction of the relationship. Pearson’s product monent correlation coefficent is used where you have variables which represent measurements of some form. Use Analyze > Correlate > Bivariate
CORRELATION Spearman’s rank correlation coefficent can also be used when you have merely ordered variables
REGRESSION Having discovered that two variables are correlated the next question might be can we model this data using a traight line? We first have to decide which variable is dependent on the other Use Analyze > Regression > Linear
MULTIPLE REGRESSION Multiple regression is used where we have more than one variable which might predict the dependent variable.