Skip to content

Instantly share code, notes, and snippets.

Show Gist options
  • Save pulipulichen/bb6c774266228b38032c17b0ab1f78a4 to your computer and use it in GitHub Desktop.
Save pulipulichen/bb6c774266228b38032c17b0ab1f78a4 to your computer and use it in GitHub Desktop.
PSPP,Regression,Statistics
看看網頁版全文 ⇨ 虛擬變項設0,1跟1,2有差別嗎? : Is There Any Difference between Setting 0,1 and 1,2 for the Dummy Variable?
https://blog.pulipuli.info/2023/06/is-there-any-difference-between-setting-and-for-the-dummy-variable.html
在迴歸分析中,兩者並沒有差別。
----
# 問題敘述 / Question
我在「那個才是影響依變項最多的自變項?以SPSS實作解釋型多元迴歸」這篇中以學生的資料進行解釋性多元迴歸分析,比較「性別」、「缺席次數」、「作業分析」、「期中考」與「期末考」對於「學期總分」的影響。
最後結論是「缺席次數」對「學期總分」的影響最大。
其中,「性別」的資料是類別類型,我用1跟2代表不同性別。
讀者統計人認為類別變項應該要轉換成虛擬變項(dummy variable),而值只能設為0跟1。
設成1跟2就變成順序變項,失去意義。
這可能是因為很多人忘記在變數檢視(Variable View)裡將類別變項的測量(Measure)改成「名義」(Nominal)。
好,話說回來,你是不是也很好奇,在解釋性多元迴歸分析中將虛擬變項設為「1跟2」或「0跟1」,兩者有沒有差別呢?就讓我們實際做做看吧。
----
# 分析工具:PSPP / Analysis Tool: PSPP。
https://www.gnu.org/software/pspp/。
這次我們的分析工具是PSPP。
這是類似於SPSS的統計軟體。
功能不及SPSS強大,但跑跑基本的統計倒是沒什麼問題。
而且PSPP採用GUN授權條款,可以免費下載使用。
支援Linux、Windows、MacOS多種版本。
- PSPP下載: https://www.gnu.org/software/pspp/get.html
這次我們用的是PSPP 1.4.1版本。
# 資料集 / Dataset
為了比較兩種虛擬變項的不同,我們準備了兩個資料集。
「性別」設為1跟2的dataset1_1,2。
- Google試算表線上檢視: https://docs.google.com/spreadsheets/d/1zczZnV4JtXlOA5qtSos7dZ1ygstqxYmVRXFtfBdv-SY/edit?usp=sharing
- dataset1_1,2.sav下載:GitHub, Google Drive, One Drive, Mega, Box, MediaFire
「性別」設為0跟1的dataset2_0,1。
----
繼續閱讀 ⇨ 虛擬變項設0,1跟1,2有差別嗎? : Is There Any Difference between Setting 0,1 and 1,2 for the Dummy Variable?
https://blog.pulipuli.info/2023/06/is-there-any-difference-between-setting-and-for-the-dummy-variable.html
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment