运用SPSS进行信度分析.doc

豆子編製 10/23/2019心理測驗SPSS上機：信度分析資料輸入Data輸入頁變項定義頁Reliability Analysis本次課題包含：如何進行各種信度分析1. 再測信度（Test-Retest Reliability）2. 複本信度（Alternate-form Reliability）3. 折半信度（Split-half Reliability）4. 內部一致性（Internal Consistency Coefficient）【計算係數】5. KR 20【壹】 再測信度（Test-Retest Reliability）某空間性向測驗有20題單選題，分別在十月與第二年四月施測同一組10名學生，以下是測驗結果，請計算信度。PersonABCDEFGHIJOct18165131516125810Apr18186161716145711 步驟一 按【Analyze】【Correlate】【Bivariate】步驟二 會出現下面的對話框，將左邊兩變項選入右邊Variables內，在Correlation Coefficients方盒內選取Pearson；在Test of Significance方盒內選取Two-tailed；勾選最下面的Flag significant correlations，之後按鍵。補充 若想呈現平均及標準差則可在按鍵前按進入下個對話框，在Statistics的方盒內選取Means and standard deviations，按繼續。Correlations紙筆計算結果：PersonABCDEFGHIJOct1816513151612581011.84.4226Apr1818616171614571112.84.8744X1 X2324288302082552561682556110X1X2 =1720N=10【貳】 複本信度（Alternate-form Reliability）某自我概念量表，有20題是非題，題本A與B 分別施測同一組10個人，分數愈高表示愈具有正向自我概念，以下是施測結果，請計算信度。PersonABCDEFGHIJForm A161214109111391612Form B1512151010121491613做法與再測信度相同，請參照。Correlations紙筆計算結果：PersonABCDEFGHIJForm A16121410911139161212.22.4413Form B151215101012149161312.62.289XA XB2401442101009013218281256156XAXB=1591N=106題的隨堂測驗施測5位學生， Y表示答對，N表示答錯，以下是測驗結果，請計算信度。PersonItem 1Item 2Item 3Item 4Item 5Item 6JoeYYYYNYSamYNNYNYSueYYNYYYPegNYNNYNGilNYNNYY【參】 折半信度（Split-half Reliability）步驟一 輸入資料步驟二 轉換資料為數字按【Transform】【Recode】【Into Same Variables】出現下面的對話框後將左邊方格內item1item6選至右邊String Variables內後點選鍵出現下列對話框後，將”N”定義為”0”，將”Y”定義為”1”後按鍵之後便會將資料轉換成下面的數字步驟三 將string的屬性改為numeric步驟四 計算奇數題與偶數題的和按【Transform】【Compute】即出現下面的對話框（此為奇數題的範例，偶數題亦然）結束後便會在spss Data Editor對話框中出現奇數題和偶數題的和步驟四 執行BivariateCorrelations紙筆計算結果. 計算兩個”半測驗”的相關StudentJoeSamSuePegGilOdds212111.4.4899Evens323122.2.7483X1X262612X1X2 =17N=5 校正相關係數為折半信度Spearmen-Brown prophesy formula 史比校正公式 (當兩個半測驗變異數相等時使用)Guttman prophesy formula 哥德曼校正公式 (當兩個半測驗變異數不等時使用)*折半信度* 折半信度也可直接使用SPSS計算 步驟一 輸入資料步驟二 按【Analyze】【Scale】【Reliability Analysis】將左邊方格內的變項依所需次序分前後半選入右邊items的方格內，在左下角的Model框中選取Split-half後按鍵，再按。Reliability* Method 1 (space saver) will be used for this analysis * R E L I A B I L I T Y A N A L Y S I S - S C A L E (S P L I T)Reliability CoefficientsN of Cases = 5.0 N of Items = 6Correlation between forms = .8729 Equal-length Spearman-Brown = .9321Guttman Split-half = .8889 Unequal-length Spearman-Brown = .9321 3 Items in part 1 3 Items in part 2Alpha for part 1 = -2.5000 Alpha for part 2 = .0000標準的折半信度算法：先求各題的難度，再依難度加以排序，再施行上述的方法透過平均值可看出其難度平均高的難度低 平均低的難度高排序由難度低到高2 6 1 4 5 3，在丟入變項時依單偶分為：2 1 5、6 4 3兩組，排列數據時前後排列。CorrelationsCorrelations* Method 1 (space saver) will be used for this analysis * R E L I A B I L I T Y A N A L Y S I S - S C A L E (S P L I T)Reliability CoefficientsN of Cases = 5.0 N of Items = 6Correlation between forms = .0000 Equal-length Spearman-Brown = .0000Guttman Split-half = .0000 Unequal-length Spearman-Brown = .0000Note # 11999The correlation between forms (halves) of the test is negative. Thisviolates reliability model assumptions. Statistics which are functions ofthis value may have estimates outside theoretically possible ranges. 3 Items in part 1 3 Items in part 2Alpha for part 1 = -.9000 Alpha for part 2 = .6923【肆】 內部一致性（Internal Consistency Coefficient）【計算係數】有5題問答題的隨測驗施測5名學生，每題問答題配分是5分，以下是施測結果，請計算信度。personItem 1Item 2Item 3Item 4Item 5Joe34435Sam43433Sue23323Peg44534Gil32433Dot32323步驟一 輸入資料步驟二 按【Analyze】【Scale】【Reliability Analysis】將左邊方格內的變項全選入右邊items的方格內，在左下角的Model框中選取Alpha後按鍵。步驟三 出現下列對話框候選取下列勾選處，後按鍵按。Reliability* Method 2 (covariance matrix) will be used for this analysis * R E L I A B I L I T Y A N A L Y S I S - S C A L E (A L P H A) Correlation Matrix ITEM_1 ITEM_2 ITEM_3 ITEM_4 ITEM_5ITEM_1 1.0000ITEM_2 .2970 1.0000ITEM_3 .7647 .5941 1.0000ITEM_4 .6860 .4330 .8575 1.0000ITEM_5 .1588 .8018 .4763 .4629 1.0000 N of Cases = 6.0Item-total Statistics Scale Scale Corrected Mean Variance Item- Squared Alpha if Item if Item Total Multiple if Item Deleted Deleted Correlation Correlation DeletedITEM_1 13.0000 6.4000 .5251 .6471 .8472ITEM_2 13.1667 5.3667 .6757 .7500 .8116ITEM_3 12.3333 5.4667 .8333 .8588 .7642ITEM_4 13.5000 6.7000 .7481 .7857 .8093ITEM_5 12.6667 5.8667 .5922 .7143 .8333Reliability Coefficients 5 itemsAlpha = .8457 Standardized item alpha = .8609紙筆計算結果Question12345Total ScoreJoe3443519Sam4343317Sue2332313Peg4453420Gil3243315Dot3232313i2 =.4722.6667.4722.2222.5833=16.1667, 2=7.4722 =2.7335 k=5 N=6【伍】 KR20有6題的隨測驗施測5名學生，Y表示答對，N表示答錯，以下是施測結果，請計算信度。Question123456AYYYYNYBYNNYNYCYNNYNNDNYNNNNENYNNYY步驟一 轉換資料Y正確為1，N不正確為0步驟二 與題四求內部致性的步驟相同Reliability* Method 2 (covariance matrix) will be used for this analysis * R E L I A B I L I T Y A N A L Y S I S - S C A L E (A L P H A) Correlation MatrixITEM_1ITEM_2 ITEM_3ITEM_4ITEM_5ITEM_6ITEM_11.0000ITEM_2-.66671.0000ITEM_3.4082.40821.0000ITEM_41.0000-.6667.40821.0000ITEM_5-.6124.4082-.2500-.61241.0000ITEM_6.1667.1667 .4082.1667.40821.0000* * * Warning * * * Determinant of matrix is close to zero: 8.426E-36 Statistics based on inverse matrix for scale ALPHA are meaningless and printed as . N of Cases = 5.0Item-total Statistics Scale Scale Corrected Mean Variance Item- Squared Alpha if Item if Item Total Multiple if Item Deleted Deleted Correlation Correlation DeletedITEM_1 2.2000 1.7000 .1400 . .2941ITEM_2 2.2000 2.2000 -.1846 . .5114ITEM_3 2.6000 1.3000 .6864 . -.0962ITEM_4 2.2000 1.7000 .1400 . .2941ITEM_5 2.6000 2.3000 -.2212 . .4891ITEM_6 2.2000 1.2000 .5833 . -.1042 R E L I A B I L I T Y A N A L Y S I S - S C A L E (A L P H A)Reliability Coefficients 6 itemsAlpha = .3273 Standardized item alpha = .3307紙筆計算結果Question123456Total ScoreAYYYYNY5BYNNYNY3CYNNYNN2DNYNNNN1ENYNNYY3Pi.6.6.2.6.2.6qi.4.4.8.4.8.4(p)(q).24.24.16.24.16.24=2.8, 2=1.76 =1.3266 k=6 N=521