研究採用主成份分析 (Principle Component Analysis)，並以最大變異法 (Varimax Method)進行直交轉軸，取特徵值(Eigenvalue)大於1的因素，進一步以因素負荷量來篩選題項，因素負荷量若小於0.5的題項不予採計。本研究依據以上條件對購屋考慮因素進行因素分析In our research, we adoptprincipal component analysis and the (Varimax Method) to do orthogonal axis,taking values of Eigenvalue factors greater than 1, then take the factorloading to correct items, if factor loading items are not less than 0.5 we willnot adopt.為了檢測問卷選項的可靠性 我們使用信度分析每一個刪除之最右側「項目刪除時的Cronbach’s Alpha值」，可發現沒有一項超過0.801，表示這些項目均不用刪除。12個項目ALPHA的值是0.801所以我們可以接受這個值To test the scaleof Questionnaire reliability we use the reliability analysisAfter analysis wefound that the item-deleted Cronbach's Alpha values have no greater than 0.801 So there is no need to remove any item from the scale.The alpha coefficient for the twelve itemsis.0.801,suggesting that the items have relativelyhigh internal consistency so it is acceptable.
我們首先要做的事 這個資料是否適合進行因素分析將所獲得之資料，再進行KMO取樣適當性檢定及巴氏球型檢定，KMO=0.850大於0.8表示這分析效果是好的，巴氏球型檢定值754.966，顯著性=0.000<=0.01，顯示資料非常適合進行因素分析。 The first decision we faces is whether or not the dataare appropriate for factor analysis.so we use the KMO measure and andBartlett's Test of Sphericity.The value of KMO measure of Sampling Adequacy for this set ofvariables is 0.850 that the correlation matrix is appropriate for factoring .
Component 2:included the variables: " house- aged"; and "parking place"; and " Feng Shui"; and "Housing type" ; and "the change of house price "; and "infrastructure"
Component 3:included the variables: " School district" ; and " administrative division" .These variables are relate to the area and location, so we called"Sector"
- H. M. CLv 79 年前最佳解答
In our research, we adopted principal component analysis and the (Varimax Method) as the values in the orthogonal axes, taking values of Eigenvalue factors greater than 1, then taking the factor loading to correct items. If factor loading items are not less than 0.5, we omit them.To test Questionnaire's reliability we performed a reliability analysis.
After analysis we found that the deleted-items' Cronbach's Alpha values are no greater than 0.801, Therefore, there is no need to remove any item from the scale.The alpha coefficient for the twelve items is.0.80, suggesting that the items have relatively high internal consistency so they are acceptable.
The first decision we face is to determine whether the data is appropriate for factor analysis. We used the KMO measurement and and Bartlett's Test of Sphericity.The value generated from KMO measurement of Sampling Adequacy for this set of variables is 0.850, that the correlation matrix is appropriate. Component 2 included the variables: "age of property", "parking space", "Feng Shui", "Housing type", "the change of house price" and "infrastructure"Component 3 included the variables "School district" and "administrative division" .These variables relate to the area and location, so we called them "Sector"