A researcher is trying to identify predictors for reading comprehension among children. Thedata filecontains 6 variables. The dependent variable is reading comprehension (reading), the independent...

1 answer below »

A researcher is trying to identify predictors for reading comprehension among children. Thedata filecontains 6 variables. The dependent variable is reading comprehension (reading), the independent variables are phoneme awareness (phoneme), visual perception (visual), morpheme awareness (morpheme), gender (1-female, 2-male) and LS (Learning Style:1-visual, 2-auditory, 3-kinesthetic).


*Please refer toSPSS Commandsfor tips on analyzing data.



Your task:



  1. Create dummy codes for categorical predictor variables. This will involve creating two new dichotomous variables from the LS variable: 1) LS_visual (where "1" indicates a visual learning style and "0" indicates not a visual learning style), and 2) LS_auditory (where "1" indicates an auditory learning style and "0" indicates not an auditory learning style). Provide a frequency table for the dummy coded variables.

  2. Check the assumptions of normality, homoscedasticity, and collinearity. Describe and provide support regarding whether the assumptions were met (Include supporting tables/graphs).

  3. Run multiple regression using three different methods (forced entry, stepwise, and hierarchical analysis). For the hierarchical analysis, consider the following variables first: gender, LS_visual and LS_auditory. Describe, compare, and interpret the results (include supporting tables). Which method would you pick for this analysis? Why?

  4. Write a summary report
    Please write a summary report to include your main findings, your interpretation to the results, critical analysis to the results, implication of the findings and conclusion.


Your assignment will be graded using quantitative reasoning rubric.


Assignment Expectations



Critical Thinking:Expressing quantitative analysis of data to support the discussion showing what evidence is used and how it is contextualized.



Interpretation:Explaining information presented in mathematical terms (e.g., equations, graphs, diagrams, tables, words)



Presentation:Ability to convert relevant information into various mathematical terms (e.g., equations, graphs, diagrams, tables, words)



Conclusions:Drawing appropriate conclusions based on the analysis of data.

Answered Same DayApr 29, 2021

Answer To: A researcher is trying to identify predictors for reading comprehension among children. Thedata...

Bolla V V Satyanarayana answered on May 01 2021
132 Votes
A Study of the Linear Relationship of children’s reading comprehension and phoneme awareness (phoneme), visual perception (visual), morpheme awareness (morpheme), gender (1-female, 2-male) and LS (Learning Style:1-visual, 2-auditory, 3-kinesthetic).
Dependent variable: Children’s reading comprehension
Independent variables: phoneme awareness (phoneme), visual perception (visual), morpheme awareness (morpheme), gender (1-female, 2-male) and LS (Learning Style:1-visual, 2-auditory, 3-kinesthetic).
The aim of the present stud
y was to investigating possible relationship between Children’s reading comprehension and phoneme awareness (phoneme), visual perception (visual), morpheme awareness (morpheme), gender (1-female, 2-male) and LS (Learning Style:1-visual, 2-auditory, 3-kinesthetic).
1) Descriptive Statistics:
Create dummy codes for categorical predictor variables.
Frequency Table:
    LS_visual
    
    
    Frequency
    Percent
    Valid Percent
    Cumulative Percent
    Valid
    Not a visual learning style
    196
    54.1
    54.1
    54.1
    
    Visual learning style
    166
    45.9
    45.9
    100.0
    
    Total
    362
    100.0
    100.0
    
Bar graph for Visual Learning Style:
Frequency for Auditory Learning Style:
    LS_Auditory
    
    
    Frequency
    Percent
    Valid Percent
    Cumulative Percent
    Valid
    Not auditory learning style
    191
    52.8
    52.8
    52.8
    
    Auditory learning style
    171
    47.2
    47.2
    100.0
    
    Total
    362
    100.0
    100.0
    
Bar Chart for Auditory Learning Style:
Frequency Table for Learning Style
    Learning Style
    
    
    Frequency
    Percent
    Valid Percent
    Cumulative Percent
    Valid
    Visual
    61
    16.9
    16.9
    16.9
    
    Auditory
    155
    42.8
    42.8
    59.7
    
    Kinesthetic
    146
    40.3
    40.3
    100.0
    
    Total
    362
    100.0
    100.0
    
2)
Test of Normality:
We check the variables Children’s reading comprehension and phoneme awareness (phoneme), visual perception (visual), morpheme awareness (morpheme) are follows normal distribution.
    Tests of Normality
    
    Kolmogorov-Smirnova
    Shapiro-Wilk
    
    Statistic
    Df
    Sig.
    Statistic
    Df
    Sig.
    Phoneme
    .178
    362
    .000
    .914
    362
    .000
    Visual
    .127
    362
    .000
    .962
    362
    .000
    Morpheme
    .211
    362
    .000
    .933
    362
    .000
    Reading
    .134
    362
    .000
    .959
    362
    .000
    a. Lilliefors Significance Correction
    
    
    
From the output
Reading Comprehension:
Shapiro Wilk Test statistic of Reading comprehension = 0.959
P value = 0.000
Here we observe that p-value(0.000) is less than level Significance value(0.05).So the variables Reading Comprehension was not follow the Normal distribution
Box-Plot for Reading Comprehension:
Interpretation: From the Box-Plot: Most of the data false on Right side. The data follows Positively skewed distribution. So the Reading comprehension is follow the normal distribution. Because The box plot is not symmetrical.
Phoneme awareness (phoneme):
Shapiro Wilk Test statistic of Phoneme awareness = 0.914
P value = 0.000
Here we observe that p-value (0.000) is less than level Significance value (0.05). So the variables Phoneme awareness was not follow the Normal distribution
Box-Plot for Phoneme awareness:
Interpretation:
From the Box-Plot: Most of the data false on Right side. The data follows Positively skewed distribution. So, the Phoneme awareness is follow the normal distribution. Because The box plot is not symmetrical.
Visual Perception:
Shapiro Wilk Test statistic of Visual Perception = 0.962
P value = 0.000
Here we observe that p-value (0.000) is less than level Significance value (0.05). So the variables Visual Perception was not follow the Normal distribution
Box-Plot for Visual Perception:
Interpretation:
From the Box-Plot: Most of the data false on Right side. The data follows Positively skewed distribution. So, the Visual Perception is follow the normal distribution. Because The box plot is not symmetrical.
Morpheme awareness:
Shapiro Wilk Test statistic of Morpheme awareness = 0.933
P value = 0.000
Here we observe that p-value (0.000) is less than level Significance value (0.05). So the variables was Morpheme awareness not follow the Normal distribution
Box-Plot for Morpheme awareness:
Interpretation:
From the Box-Plot: Most of the data false on Left side and there is one Outlier that is 0.00. The data follows Negatively skewed distribution. So, the Morpheme Awareness is follow the normal distribution. Because The box plot is not symmetrical.
Test of Homoscedastic
Collinearity Test:
    Coefficients
    Model
    Unstandardized Coefficients
    Standardized Coefficients
    T
    Sig.
    Collinearity Statistics
    
    B
    Std. Error
    Beta
    
    
    Tolerance
    VIF
    1
    (Constant)
    .005
    .059
    
    .086
    .931
    
    
    
    phoneme
    .623
    .056
    .499
    11.222
    .000
    .792
    1.262
    
    visual
    .236
    .049
    .214
    4.842
    .000
    .802
    1.247
    
    morpheme
    .143
    .050
    .118
    2.850
    .005
    .914
    1.094
    
    Gender
    -.028
    .021
    -.053
    -1.340
    .181
    .985
    1.016
    
    Learning Style
    -.012
    .014
    -.033
    -.821
    .412
    .982
    1.018
    a. Dependent Variable: Reading
b.
    
    
    
    
    
    
    Collinearity Diagnostics
    Model
    Dimension
    Eigenvalue
    Condition Index
    Variance Proportions
    
    
    
    
    (Constant)
    Phoneme
    visual
    morpheme
    Gender
    Learning...
SOLUTION.PDF

Answer To This Question Is Available To Download

Related Questions & Answers

More Questions »

Submit New Assignment

Copy and Paste Your Assignment Here