ED 602 Statistical Research for Behavioral Sciences Brian G. Smith, Ph.D. |
|
Definitions Archival records research - answering scientific questions from information in existing records. If we were basing our measure of student behavior on information in their permanent school records, that would be archival records research. Meta analysis and historical research are two popular forms of archival records research. Bar graph - a graph used to present a frequency distribution for qualitative data Class interval - the range of score values into which the raw scores are grouped in a grouped frequency distribution Constant - a value that does not change for the duration of a study. Gender is usually constant during a study. Continuous Variable - A variable that can assume an infinite set of values between any two levels of the variable. A studentís height can be measured with ever increasing accuracy (1.7 meters, 1.73 meters, 1.732783956 meters, etc.) so it is continuous. Correlational method - A type of research in which two or more variables are measured but not manipulated, and the relationship between the variables is assessed. If we were looking for a relationship between BASC scores and absenteeism, we would be doing correlational research. Cumulative frequency of a score -( cf ) - the frequency of occurrence of a score plus the sum of the frequencies of all the scores of lower value. Data - the scores or measurements obtained by a scientist doing research. This can be test scores, survey answers, observations made, anything recorded by the scientist for later analysis. Dependent variable - The variable that is measured in an experiment. If we are testing the effects of high and low sugar diets on child behavior, then we watch to see if the groups will have different levels of misbehavior once the study begins. The misbehavior of each group in response to the new diets, is the dependent variable. Discrete Variable - A variable that can be measured only with a finite set of values. The BASC scores are given in whole numbers, making it discrete. Empirical data - scores or measurements based on sensory experience or observation. Responses to the BASC survey are empirical, because we can see or observe them. Experimental method - manipulating one or more independent variables in a carefully controlled situation. Randomly assigning students to one of two groups and teaching one group with a new spelling method, and the other group with the traditional method would be experimental research. Frequency distributions - A table showing each score in a set of scores and the number of times it occurred
Frequency polygons - Connected dots indicating the frequency at the midpoints of classintervals with straight lines Histograms - a bar graph in which size of the class interval is represented by the width of the bar on the abscissa (x axis). and the frequency of scores in the class interval is given by the height of the bar. Independent variable - The variable that is manipulated in an experiment. If we are testing the effects of high and low sugar diets on child misbehavior, then the diet is what changes for each group of students being studied. Diet is the independent variable. Interval Scale - Assigning numerical values to a variable with an arbitrary zero point. A zero on the BASC survey does not mean that the student has an absence of behavior, so it is measured on an interval scale. Measurement - Assigning numbers to variables following a set of rules. The BASC comes with a strict set of rules for scoring, or measuring, student responses. Measures of Central Tendency - Numbers that represent the average or typical score obtained from measurements of a sample. Mean, median and mode are the three most common measures of central tendency. Measures of Variability - Numbers that indicate how much scores differ from each other and the measure of central tendency in a set of scores. These will be covered in detail in later chapters, but variance and standard deviation are two examples of measures of variability. Median - A score value in the distribution with an equal number of scores above and below it. The median is the 50th percentile in a distribution. Mode - the most frequently occurring score in a distribution
Naturalistic Observation - observing behaviors occurring in natural settings without intruding into the situation. Portions of the BASC that are not being used in our study involve observing the students in the natural setting of their regular classroom. This is also referred to as ethnographic research. Nominal Scale - Classification of a measured variable into different categories. We can classify our students as 0 for males and 1 for females. The numbers have no meaning other than to differentiate between the two genders. Operational definition - specifies the procedures used to manipulate an independent variable or to measure a dependent variable. The operational definition for our study is that we are giving written copies of the BASC student self-report survey to our subjects in the controlled environment of the counselorís office. Ordinal
Scale - Arranging characteristics of a variable along
an ordered continuum from largest to smallest. If we are also interested
in the class ranking of our students, their ranking would be measured
on an ordinal scale. Parameter - a single number used to describe a characteristic of a population, often symbolized by a Greek letter. A parameter for 4th grade students might be that the average 4th grade student is 9 years old. Percentiles - a score at or below which a specified percentage of the scores ina distribution fall Percentile rank - the percentage of scores in a distribution that are equal to or less than that score Population - the complete set of people sharing the common characteristic specified by a researcher. So, if a study is on fourth grade students, then the population is ALL the fourth grade students in the world. If a study is on freshmen at MSUM, then the population is ALL the freshmen in MSUM. Population mean () - The sum of all the scores in a population divided by the number of scores summed. This is the same as a sample mean, but using every score for every subject, rather than just a small sampling. Qualitative data - Nominal measurements, which categorize the measured variable. Again, the gender of our students would be qualitative data. Quantitative data - Measurements that provide numerical information about the variable measured, such as the BASC scores. Quasi-Experimental method - A loosely defined type of research that is intended to be run as an experiment, but with limitations to the randomness of the sample. If we tested all the Hispanic students in one English as a Learned Language class and all the Caucasian students in a regular English class rather than drawing names from all the students in the school, we would have a quasi-experimental method. Range– The numeric difference between the lowest and the highest scores in a distribution. Raw data - The scores of a subject exactly as collected and before they areanalyzed statistically Ratio Scale - Assigning numerical values to a variable with a scale that possesses a physically real zero point. The studentsí heights, while irrelevant, are measured on a ratio scale. Research hypothesis - the predicted relationship between an independent and a dependent variable. If we are testing the effects of high and low sugar diets on child misbehavior, then we might have as our research hypothesis, "There will be no significant difference in child misbehavior between the low sugar and high sugar groups." Research method - A general approach used by a behavioral scientist to collect data. We will be using the survey method for collecting data on our two groups of students. Sample - a subset of a population. A sample of our 4th grade population might be the 4th grade students in Mr. Bryantís classroom, or the 4th grade students in Moorhead schools. Sample mean ( ) - The sum of a set of scores divided by the number of scores summed. In adding all our BASC scores I get 1285, then I divide by 25 (the number of scores) and I get a sample mean of 51.40. Statistic - a single number used to describe data from a sample, often symbolized by a roman letter. A statistic for Moorheadís 4th grade classes is that the average age is 9.5. (Note that a statistic is just like a parameter, except for the group of people being described.) Stem and leaf display - A display of data in which the first digit of a score is the stem, and the last digit is the leaf. Skewed - when a distribution has scores clustered more at one end than at the other. Standard deviation – a measure of variability that represents an average of how much scores vary from the mean. Sum of squares (SS) - A numerical value obtained by subtracting the mean of a distribution from each score in the distribution, squaring each difference, and then summing the differences. This number by itself means very little, but it is a key component of many statistical calculations. Survey method - obtaining data from oral or written interviews with people. Our study involves the student self-report survey portion of the BASC, so we are doing the survey method. Symmetrical frequency distribution - A distribution in which one side is the mirror image of the other side Variability– How much scores differ from each other and the measure of central tendency in a distribution. Variable - Any environmental condition or event, stimulus, personal characteristic or attribute, or behavior that can take on different values at different times. For example scores on a test can change (think pretest/posttest), weight can change in response to diet or exercise, opinions can change, age can change in longitudinal studies, or teaching methods can be changed. Variance– a measure of dispersion that produces results in terms of square units, and for that reason rarely used. |