Correlation

Emile<

Reviewed Question

  1. Correlation studies are designed to investigate the nature and strength of functional relationships among the variables of interest to the researcher.
  2. (1)Multiple regression and Multiple R (2) Two-way ANOVA, three-way ANOVA, etc. (3) Friedman's two-way ANOVA
  3. Ordinal type-one independant variable and Ordinal type-one depedant variable

  4. H0:r=0 There is no systematic relationship between the dependent and independent variables.
    H1:r>0 There is positive relationship between the dependent and independent variables.
    H2:r<0 There is negative relationship between the dependent and independent variables.
  5. Answer is 100+10X2=100+20=120
  6. (1)10(2)100(3)2(4)120

  7. (1)each pair of scores is independent from all other pairs;
    (2)each of the two variables involved is normally distributed;
    (3)each of the two variables involved is measured on an interval scale;
    (4)the relationship between the two sets of scores is linear.

  8. (1)Look at the hypotheses.
    (2)Look at α level;
    (3)Compare the observed and critical statistics.
    (4)a.If the observed statistic is less than the critical statistics, accept the null hypotheses and stop.
    b.If the observed statistic is greater than the critical statistic, reject the null hypothesis and continue.

    (5)Decide which alternative hypothesis is more logical.
    (6)Interpret the results in terms of the p level.
  9. The standard error of estimate (SEE) is an estimate of the margin of error in real score points-usually Y score points.

  10. a.restriction of range:The range has been restricted, and this will tend to make any resultant correlation coefficient lower.
    b.skewedness: There simply is not room for scores to be distributed two standard deviations below the mean.
    c.causality.: one variable, although highly related to another, may not be "causing" it.

Application

I choose "Standardizing Infra-red Measures of Bone Mineral Crysallinity: and Experimental Approach" by Todd A. Surovell and Mary C. Stiner from Journal Archaeological Science(2001) as my analytical article. Their hypothesis is if grindling is overly intensive, SF (Splitting factor)measurement may respond more to variation in sample preparationthan to the differences in bone mineral crystallinity. Their conclusion is that because even slight differences in sample preparation can affect SF values, a set of procedures and standards are proposed as a means of calibrating infra-red crystallinity measures. They use scatterplot to predict the possible result. This chapter help me to understand the correlation which may just a relation, not casuality tograsp what the authors want to express the goal of the experiment.

ActivStat

ACT-1


  1. (a)Graph->Interactive->Scatterplot->Put time in Y axis and put size in X axis.(b)There is no obvious relation, whether positive or negative, between the size and time.
  2. (a))Graph->Interactive->Scatterplot->Put time in Y axis and put distance in X axis.(b)There is no obvious relation, whether positive or negative, between the distance and time.
  3. (a))Graph->Interactive->Scatterplot->Put time in distance axis and put size in Y axis.(b)I still can not find any trend between distance and size.
  4. From the size-time,distance-time and distance-size pair, the scatterplot disply nonunidirectional distributed. I suggest that we must see the original distribution of all circles, because the coordinal axis may the possible factors to affect the results.
  5. The correlations of time-size and time-distance are -0.304 and 0.355, and they are so similar. But in distance and size, it is 0.009 and so differenct from previous pairs. This means size and distance affect time separately.

Family economic


(a)Graph->Interactive->Scatterplot->Put income in Y axis and put food in X axis.(b)There is positive relation between the food and income.

Progress Evaluation

  1. The teacher,classmate and the textbook. Because I can learn the unfamiliar discipline happily.
  2. (1)the class time.Because I find it is so tired for me in the 13:30.(Sorry it is my problem.)
    (2)The difference between textbook and ActivStat. It oftern confuse me.
  3. For (1), I may change my daily life, and for (2) I consider we can notice the difference to explain what it is cuased.
  4. The learning technique.
  5. Listening ,speaking and writing ability in English.
  6. I will manage to speak more to improve the shortcoming.
  7. Sorry, I still need more time to find the answer.
  8. I have some learing problem in other courses, but I still can not understand it clear. I want to use CM in archaeological research.