This method can help answer key questions in the social sciences field about the psychology of reading and language development. This technique illustrates how individual difference characteristics co-vary and how a unique effect may pail in comparison to an effect shared with other characteristics. The implications of this technique extend to the social sciences more broadly.
As it allows accounting for the commonality of predictors that help us understand overlapping variance. To read the data into the software with a graphical user interface, click file and hover the mouse over open. Click data and locate the relevant data file on the computer.
If the file type is not consistent with software with a graphical user interface, click files of type, and select the appropriate file format. Then click open. To explain the total variance based on two independent variables, click analyze and hover the mouse over regression to select linear.
Click on the dependent variable in the variable list, followed by the arrow next to dependent. Click on the two independent variables in the variable list and click the arrow next to independence. Click okay and click the viewer window of the software.
Use the mouse to scroll to the model summary section and record the value under R square. Then label this value total R squared. To explain the total variance based on independent variable one, repeat the total variance explanation with independent variable one only in the independent variable list and click the viewer window of the software.
Then use the mouse to scroll to the model summary section, record the value under the R square column, and label this value independent variable one R squared. To explain the total variance based on independent variable two, repeat the total variance explanation with independent variable two only in the independent variable list and click the viewer window of the software. Then use the mouse to scroll to the model summary section, record the value under the R square column, and label this value independent variable two R squared.
To compute the unique, common, and unexplained variance components, open the data management software, and enter the total R squared independent variable one R squared and independent variable two R squared into cells A1, B1, and C1 respectfully. Enter the total R squared value into A2.The independent variable one R squared value into B2, and the independent variable two R squared value into C2.To calculate the unique variance of variable one, enter the formula into D2, and label this value as unique variance of variable one in D1.To calculate the unique variance of variable two, enter the formula into E2, and label the value as unique variance of variable two in E1.To calculate the common variance between variables one and two, enter the formula into F2, and label this value as common various between variables one and two in F1.To calculate the unexplained variance, enter the formula into G2, and label this value unexplained variance in G1.To plot the unique variance of variable one, the unique variance of variable two, the common variance between variables one and two, and the unexplained variance, click and drag the mouse cells over D2, E2, F2, and G2 to highlight the data, and click insert, then click charts, pie chart, and 2D pie chart. In this representative study of unique and common variances of language and decoding for predicting reading comprehension, the regression analysis for grade one students accounted for 60%of the total variance in reading comprehension.
When the variance in grade one was decomposed into unique and common effects, decoding uniquely explained the 24%of variance in reading comprehension, and language uniquely explained the 17%of variance. The common variance of decoding and language was 19%In grade seven, the regression analysis accounted for 53%of the total variance in reading comprehension. With decoding uniquely explained by 7%of the variance in reading comprehension, and language explained 28%of the variance.
The common variance of decoding and language in explaining the variance in reading comprehension was 18%In grade 10, the regression analysis accounted for 61%of the total variance in reading comprehension. With decoding uniquely accounting for a 6%of the variance, and language uniquely accounting for 42%of the variance. The common variance of decoding and language in explaining the variance in reading comprehension was 13%While attempting this procedure, it's important to remember that the decomposition process differs from typical regression approaches of separately calculating the unique variance that each independent variable explains.
This protocol can be modified to answer additional questions about whether the proportions of unique and common variances differ by socioeconomic status. Further, the observable variables can be replaced with latent variables to reduce measurement error. The amount of common variance that decoding and language together explain in predicting reading comprehension, especially in the elementary grades suggest that instruction should focus on the integration of linguistic knowledge at the word level.
