Advanced Statistics Portfolio

The portfolio should be completed independently. (Ie., you may not work together on this assignment).

The portfolio assignment assesses your ability to choose and apply the correct statistical test for the research questions asked. Explain how you would address each research questions using the primary school data set (unless asked otherwise*) provided on Blackboard. You should illustrate your answer making appropriate but not superfluous use of software output so that following points are clear in each answer.

1. The preliminary analyses you would conduct and what you would be looking for.
2. Which statistical test you would run.
3. How you would interpret the results (e.g, which statistics you would report and what would they tell you?)

Hint: All the BAS assessments are standard scores which reflect a child’s abilities relative to their age group. These standard scores have a mean of 50 and a standard deviation of 10. Explanation of all the variables included in the data sets has been provided in lecture and is also available as a handout on Blackboard.

1. [6 points] You are a researcher interested in children’s literacy skills, measured with a set of standardised assessments of single word reading (BASR) and spelling (BASSp). Create a new variable that represents the average of reading and spelling performance in the sample. Using your knowledge of data screening procedures, answer the following questions for your new variable.
   1. Provide descriptive statistics of central tendency and dispersion on this new variable [3 pts]
   2. Is the variable normally distributed? Give at least one graphical and one statistical test [3 pts]
2. [12 points] Evaluate the distributions for the reading time measures for regular, irregular and nonword reading (CCRT, CCIT, CCNT)
   1. Provide descriptive statistics of central tendency and dispersion on these variables [3 pts]
   2. Are the variables normally distributed? Give at least one graphical and one statistical test [3 pts].
   3. If not, apply an appropriate transformation and test whether the transformed variable is normally distributed [3 pts].

1. Which transformation worked the best for each variable? [3 pts]

1. [12 points] Dyslexia is usually defined as a deficit of literacy achievement of at least 1.5 standard deviations from average. Using your literacy skill variable created in part 1, above, create a categorical variable that can be used to classify each participant in the data set as either a dyslexic or control.
   1. How many children are classified as dyslexic using this technique? [3 pts]
   2. Is there an association between gender and dyslexia? [3 pts]
   3. Using the BAS Digit span (BASDS) measure of short term memory, evaluate whether the children with and without dyslexia differ on this measure. [3 pts]
   4. Evaluate the effects of dyslexia and gender, accounting for the effects of age on BASDS. [3 pts]
2. [24 points] You want to find out whether children’ performance on the Spoonerisms measure varies as a function of reading age, digit span and orthographic skills.
   1. You have collected multiple measures of Spoonerisms and orthographic skills and want to know whether these measures of the same construct are being answered in the same way. State a test that can be performed to answer this question and perform this analysis [4 pts]
   2. Having performed the above tests, evaluate whether or not you are justified in combining the items into average measures of Spoonerisms and orthographic skills. [2pts]
   3. Compute these average variables. [2 pts]
   4. Before regressing Spoonerisms on orthographic skills, reading age and digit span, you want to be sure that the independent variables are a) significantly correlated with the dependent variable and b) the independent variables are not too highly correlated with each other. Identify and explain what problem occurs if IVs are too highly correlated. [4 pts]
   5. Correlate Spoonerisms, orthographic skills, reading age and digit span. Report the results and comment on the correlations found, in particular stating whether you think it is acceptable to include all of the variables in a single regression. [6 pts]
   6. Regress Spoonerisms on orthographic skills, reading age and digit span using standard regression. Report the results and state which variable(s) significantly predict Spoonerisms and which variable is the best predictor. [6 pts]
3. [10 pts] You want to find out whether dyslexia is associated with lower scores on Spoonerisms, but you are also aware that Spoonerisms scores may also differ according to the age of the participants. Assess the significance of the difference in Spoonerisms scores for dyslexics and controls in equal sized groups of younger and older children.
   1. Is there a significant effect of age on Spoonerisms score? [6pts]
   2. Is there a significant effect of group on Spoonerisms score? [2pts]
   3. Is there a significant group x age interaction effect for Spoonerisms score? [2pts]
4. [12 Points] You want to find out how well the following variables in the study predict dyslexia in the sample: Spoonerisms (total score), CCI accuracy, CCN accuracy, Orthographic skills (total score from Ols1 and Ols 2), gender and age (months). Choose and appropriate method of analysis.
   1. How well does the set of variables predict group membership? [6 pts]
   2. Which of the variables in the set is the best single predictor? [6 pts]
5. [12 points] You want to evaluate whether the measures of phonological and orthographic skills used in the study actually measure these independent constructs. Using the following variables of phonological skill (CCN accuracy, CCN time, Spoonerisms total score) and orthographic skill (CCI Accuracy, CCI time, Ols total score), evaluate whether these measures actually assess different aspects of literacy skill.

Price: $33.21

Solution: The downloadable solution consists of 18 pages, 1521 words and 24 charts.
Deliverable: Word Document