Reflection is required for EdTech 562, Introduction to Statistics for Educational Technology. I’m finding it hard to reflect on the topic and feel that an initial first step involves summarizing the content. So, this is a summary more than a reflection of the current module I’m working through in EdTech 562.
Social researchers state their ideas based on perceived notions in the form of an hypothesis. Hypotheses indicate a relationship between or among two or more variables. Variables are identified as independent when representing the presumed cause and dependent when representing the presumed effect of the relationship. Researchers assumptions about the cause and effect of identified variables are clearly stated in the hypothesis. Data collection provides systematic evidence for the hypothesis. Statistics involves the organization and classification of quantifiable data providing evidence for or against an hypothesis. Terms used by researchers to help define data in order to describe and infer assumed relationships are: nominal (categorical), ordinal (ordering of categories), and interval (scale with equal measures).
Organization and classification of quantifiable data involves developing frequency distributions to determine proportions and percentages. When data are lengthy frequency distributions are grouped into class intervals. Using the mid-point of a class interval is sometimes helpful when a representative score for all data in a class interval is needed. Frequency distributions are typically represented in tables and graphs. In general, tables include columns representing categories, mid-points (when applicable), frequencies, and percentages laid out from left to right respectively. At times cross-tabulations are used to explore the relationships of multiple variables placing variables in columns and rows. Generally, with cross-tabulations, percent calculations are of the independent variable.
Graphical representations of collected data are often displayed in addition to tables. Pie charts easily display categorical data that adds up to 100%. Bar graphs and histograms are more versatile and frequently used by social researchers. Ordinal and interval data are frequently displayed using a frequency polygon. The skewness of asymmetrical distributions and peakedness (kurtosis) of symmetrical distributions are significant to social researchers. Trends over time are often displayed with line graphs.
SPSS is a software program for data analysis. It can streamline the process of screening and cleaning data. Clean data is of considerable importance to a researcher. SPSS can quickly analyze data to find errors and allows for error correction. Screening and cleaning data is a critical first step to data analysis.
References
Levin, J. & Fox, J. A. (2011). Elementary statistics in social research: The essentials. Boston: Pearson Education, Inc.
Pallant, J. (2013). SPSS survival manual: A step by step guide to data analysis using IBM SPSS. New York, NY: McGraw Hill.
Susan Shannon: EDTECH Learning Log 