DOWLING COLLEGE: School of Education Department of Educational Administration, Leadership, and Technology

Size: px

Start display at page:

Download "DOWLING COLLEGE: School of Education Department of Educational Administration, Leadership, and Technology"

Sibyl Hines
5 years ago
Views:

1 1. Doe 2. Doe 3. Doe 4. Doe DOWLING COLLEGE: School of Educatio Departmet of Educatioal Admiistratio, Leadership, ad Techology 5. Calculate meas ad stadard deviatios for per capita icome ad total reveues per studet (for the year). Statistics PER CAPITA INCOME (CENSUS) [ ] N Valid Missig 2 2 Mea Media Mode a 5370 a Std. Deviatio Percetiles a. Multiple modes exist. The smallest value is show For each variable, determie which states are at the 25 th ad 75 th percetile. Per Capita Icome: 25 th % = Alabama with a z-score of th % = Washigto with a z-score of.68 Total Reveues per Studet: 25 th % = North Carolie with a z-score of th % = Maie with a z-score of.55 Where does New York lie? Per Capita Icome = 84.3% with a z-score of.81 Total Reveues/studet = 96.1% with a z-score of 2.21 Calculate its z-score for each variable [(x-µ)/σ] ad iterpret this umber i oe setece for each variable. The z-scores for NY are clearly high as a result of the states per capita ad total reveue/studet beig much more tha most others i the uio. The NY per capita is almost oe SD above the rest but it more tha 2 SD s above the mea.

2 6. Did pupil-teacher ratio or total reveues per studet differ sigificatly from the school year to the school year? Calculate ad explai. Paired Samples Test Paired Differeces Pair 1 Pair 2 PUPIL/TEACHER RATIO (STATE) - PUPIL/TEACHER RATIO (STATE) [ ] PER STUDENT [ ] - PER STUDENT [ ] Std. Std. Error 95% Cofidece Iterval of the Differece Mea Deviatio Mea Lower Upper t df As highlighted above, there is a clear sigificace betwee pupil-teacher ratio ad total reveues per studet from the school year to the school year 7. What is the relatioship betwee the followig variables (either from the school year or the year 2000): pupilteacher ratio, total reveues per studet, per capita icome, media icome male, media icome female, percet below poverty, percet free ad reduced luch, percet LEP/ELL studets, ad percet Idividualized educatio program (IEP) studets? Provide a chart ad as briefly as possible, describe sigificat relatioships. PUPIL/TEACHER RATIO (STATE) PUPIL/TEACHER RATIO (STATE) PER CAPITA INCOME (CENSUS) s MEDIAN INCOME- MALE (CENSUS) MEDIAN INCOME- FEMALE (CENSUS) PERCENT_LEP/ ELLSTUDENT PERCENT_FREEa dreduced- Luch_1999_00 PERCENT INDIVID.EDUC.P ROGRAM STUDENTS TOTAL POP.- BELOW POVERTY LEVEL (CENSUS) ** ** ** **.704 **.681 ** *

3 PER CAPITA INCOME (CENSUS) MEDIAN INCOME- MALE (CENSUS) MEDIAN INCOME- FEMALE (CENSUS) PERCENT_LEP/ELL STUDENT PERCENT_FREEad Reduced- Luch_1999_00 PERCENT INDIVID.EDUC.PRO GRAM STUDENTS TOTAL POP.- BELOW POVERTY LEVEL (CENSUS) AJ Hepworth EDE 9803 S. Marshall Perry, Ph.D ** **.935 ** * **.874 ** ** **.356 * **.935 **.820 ** ** ** N * * ** * N * N **.347 * As highlighted above, several sigificat relatioships exist ad two almost sigificat relatioships exist. I would ecourage lookig agai usig other correlatio aalysis methods suchas Spearma to see if they ca become more sigificat.

4 8. Usig the variables above, how well ca a model predict pupil teacher ratio? Provide a chart to summarize the best model (hit: stepwise) ad explai. Model Summary Model R R Square Adjusted R Square Std. Error of the Estimate a b c a. Predictors: (Costat), PERCENT_LEP/ELLSTUDENT b. Predictors: (Costat), PERCENT_LEP/ELLSTUDENT , c. Predictors: (Costat), PERCENT_LEP/ELLSTUDENT ,, MEDIAN INCOME- MALE (CENSUS) The three geerated models above show fair R-square models with model 3 beig he best with a R square score of.586. The factors to best predict pupil teacher ratio are: PERCENT_LEP/ELLSTUDENT ,, MEDIAN INCOME- MALE (CENSUS)

5 9. For the school year, what are the relatioships amog the followig variables: pupil-teacher ratio, total reveues per studet, local reveue percet of total reveue, state reveue percet of total reveue, ad federal reveue percet of total reveue? s PUPIL/TEACHER RATIO (STATE) [ ] PER STUDENT LOCAL REV- PCT TOTAL REV (STATE- FIN.) STATE REV- PCT TOTAL REV (STATE- FIN.) FEDERAL REV- PCT TOTAL REV (STATE- FIN.) PUPIL/TEACHER RATIO (STATE) LOCAL REV- PCT TOTAL REV STATE REV- PCT TOTAL REV FEDERAL REV- PCT TOTAL REV ** * ** * * *.322 * ** ** * N * * **. is sigificat at the 0.01 level (2-. *. is sigificat at the 0.05 level (2-. Several relatioships exist above, icludig both direct ad idirect moderate relatioships. Oe egative relatioship is extremely strog, betwee state ad local reveue.

6 10. Usig the school year variables above, how well ca a model predict total reveues per studet? Provide a chart to summarize the best model ad explai. Two models were geerated both showig ear perfect predictios with R squared values of.986 or.987. It is clearly evidet that the total reveues per studet Model Summary Adjusted R Model R R Square Square Std. Error of the Estimate a b a. Predictors: (Costat), [ ] b. Predictors: (Costat), [ ], FEDERAL REV- PCT TOTAL REV 11. (Bous questio o pealty if skipped) Usig the data set as it is curretly cofigured, is it possible to determie if the media icome for females differs sigificatly from the media icome for males? If it is possible, provide relevat output ad explai. If it is ot possible, explai how the data set would eed to be cofigured (set up) for such a aalysis. No, you caot determie the media icome for males ad females based o the curret cofiguratio because all states are idepedetly listed with their respective icomes. I will aalyze the variables separately ad calculate their sums ad media value, as show below. Statistics WOW! What a differece! MEDIAN INCOME- FEMALE (CENSUS) N Valid 51 Missig 2 Media Sum Statistics MEDIAN INCOME- MALE (CENSUS) N Valid 51 Missig 2 Media Sum

ii. Interval estimation:

ii. Interval estimation: 1 Types of estimatio: i. Poit estimatio: Example (1) Cosider the sample observatios 17,3,5,1,18,6,16,10 X 8 X i i1 8 17 3 5 118 6 16 10 8 116 8 14.5 14.5 is a poit estimate for usig the estimator X ad