Methods in Research Fall/2015

Guttman-type items

(also called cumulative items) provides a series of items on a continuum linked in a manner such that a participant's response to one item in the series indicates approval of all choices below the indicated level. To be analyzed, these data must be coded in a sequence (e.g. 1,2,3, or 4 according to the item number selected) rather than each item coded separately. Somewhat complex to code, analyze, and interpret.

Parametric tests (Subdivision of Inferential Statistics)

(e.g. t-tests and ANOVA) Used to calculate hypothesis when dependent variable is measured with an interval or ratio scale and when certain other assumptions are met (e.g. normally disturbed scores, homoscedasticity). Under normal circumstances, parametric tests are more powerful than nonparametric tests.

Nonparametric tests (Subdivision of Inferential Statistics)

(e.g., chi-square, Mann-Whitney U) are used to evaluate hypotheses about the shapes of distributions and are only applied to nominal or ordinal data.

Box--and-whisker plot

(sometimes shortened to boxplot) is useful for displaying trends in data. Have to know the lowest score, highest score, median score, first quartile demarcation (Q1), and the third quartile demarcation (Q3). The box begins at (Q1) and ends at (Q3). The center line through the box falls at (Q2 the median). The whiskers are the lines protruding from the box all the way down to the lowest score and to the highest score. The primary advantage of this is the visual display of potential skewing. 50 percent of the distribution's scores fall within the box, 25 percent below the box, and 25 percent above the box.

Postmodern paradigms

A group of theories that include constructivism, critical theory, and feminism, are often used in qualitative approaches to research. View researcher as an instrument in qualitative inquiry, a tool that is changed by the research process. They conceptualize the researcher-informant relationship as a necessary component of data analysis in that researchers' interactions serve to develop and confirm findings with informants. When presenting results, are more likely to use first person rhetoric, present data in a narrative format and provide detailed descriptions regarding research bias and reflexivity.

Positivism

A paradigm closely aligned with the hypothetic-deductive method, focuses on hypothesis testing of existing theory. Seek to explain phenomena through cause-and-effect. By drawing a clear line between the researcher and the study, greater objectivity then empiricism. Designed to decrease subjectivity. Maintains that values do not belong in research. May employ methods that involve structure and manipulation to hep explain a phenomenon in a generalized manner. Attempts to remain emotionally neutral from the data, and rhetoric is objective.

Empiricism

A philosophical movement that claimed all knowledge came from experiences that can be tested against the experiences of others and not from the more personal senses. Shows external validation of findings which means it applies to others outside the current setting.

Frequency distributions

Allows researchers to quickly view the frequency with which each score occurs in a distribution; there are two types: simple and group

Curves

An extension of the frequency polygon. Start as frequency polygons but undergo a smoothing process and so the data are displayed not as sharp peaks and valleys but as a smooth progression between values. Frequently described by using the following characteristics: mean (central tendency), standard deviation (a measure of dispersion), skew, and kurtosis. The mean is the mathematical average, or "balancing point," of a distribution scores. The standard deviation is the dispersion or spread of scores around the mean. Skew describes the symmetry of a curve, and kurtosis the "peakedness" of a curve

A product

Anything created by a participant representing a program goal (e.g., artwork, composition, or poster)

Independent Variable

Are those believed to affect the behavior or status of another variable. Any counseling session is an independent variable in the life of the client because it is intended to affect client behavior. All treatments, including counseling, medications, interventions, training programs, even books, often act as independent variables in our lives. They intend to change us in some way, to cause us to be different.

Postpositivism

Asserts that researchers can only approximate a universal truth. Acknowledges that theories are meant to be tested for exceptions and disproven and that research should rigorously attempt to disprove a phenomenon to strengthen our understanding of it. Maintains that values do not belong in research. May employ methods that involve structure and manipulation to hep explain a phenomenon in a generalized manner. Attempts to remain emotionally neutral from the data, and rhetoric is objective.

Likert-type scale items

Assess a participant's attitude or preference using a question or stem accompanied by an order of response choices (e.g. Strongly Disagree, Disagree, Agree, or Strongly Agree). The classic Linket scale did not have a neutral middle choice. but researchers frequently add a middle choice (thus the term, Likert-type item). Usually researchers do not provide a numerical rating with the verbal description but add one during coding or data input t allow quantification of response choices (0 = Strongly Disagree, 1 = Disagree, etc.) It is also essential that every point on the response continuum be labeled.

Program records

Can be helpful resources of evaluation data and usually easy to locate because they are generated by organizational bureaucracies.

Organismic Variable

Characteristics of an individual that cannot be directly manipulated. Sex is organismic variable because (with few exceptions) participants can be categorized as either male or female. Researcher cannot "randomly assign" a sex to a participant, but they can categorize or group all participants as males or females and then randomly assign females and then males to various other treatment conditions.

Constructivism (postmodern concept)

Concerned with social justice and the understanding and application of context to the creation of knowledge. Critique earlier paradigms because they believe reality should never be labeled as objective, as the voice of the researcher and the informants are biased and seated in different cultural experiences and identities. As pure objectivity is unattainable; a singular "truth" does not exist.

Hallmark of Scientific Inquiry - Replication

Conducting the study over and over again. If the findings from the studies do not match, one cannot assume that new knowledge has been created, and "it's back to the drawing board." Researchers always review a study to determine how much subjectivity was allowed in, whether in the data-gathering methods or the interpretation of the results.

Hallmark of Scientific Inquiry - Objectivity

Connors that the hand and mind of the researcher are removed as far as possible from the research itself. The thoughts, opinions, and expectations of the researcher do not influence the findings that result from the data-gathering process; otherwise, the result would be a very biased definition of knowledge.

Hallmark of Scientific Inquiry - Generalizability

Define the quest for the greatest good for the greatest number of people. It is not uncommon to ask, whether a particular treatment that was tested only on adults will work equally well on adolescents. In trying to find the greatest good for the greatest umber of people, we consistently want to know what works and whether it will work on a different population.

Outcome Variable/Criterion Variable

Dependent variable is also referred to as these (in correlation research)

Response Variable

Dependent variable is also referred to as this (in behavior research)

Descriptive Statistics

Describe the summarized data.

Discussion/Interpretation Section

Detailed summary of the probably explanation(s) of the observed phenomena, trends, and anomalies are presented. Researchers try to make sense of the data, discuss the significance of the results, suggest how the results may relate to previous studies, and report any possible limitations. Researchers should always refer back to their initial hypothesis in a discussion of the results. In drawing conclusions, it is extremely important that the researchers draw only conclusions that are supported by the data. Objectivity, generalizability, and replication are all affected when the researchers "goes beyond" the data and makes unsubstantiated conclusions.

Nonmaleficence

Do no harm

Feminism (Postmodern concept)

Emphasizes the role of affect within phenomena, the researcher=participant relationship, and the political implications of research. Goal of empowerment, social action, and social justice. Believe that research values or bias should influence research process and outcome, thus inviting social activism and contributing to social justice.

Traditions and faith

Help us construct knowledge that gives us a sense of confidence and rust in ways very similar to sensory experiences. How one is raised, in what culture, and with what religious practices provide foundations for knowledge so powerful that one often cannot discern their influences on how one behaves and acts in daily life.

Inferential Statistics

Helps predict the probability of occurrence of some casual event or association with some degree of confidence and allow this prediction to be generalized back to the population from which the sample is drawn. Used to determine if one counseling approach is more effective than another on a given sample. The result may allow the counseling profession to improve the overall quality of client care when this sample result is generalized and applied to the population of all clients receiving clinical care.

Stimulas Variable

In behavior studies, the independent variable is sometimes referred to as this.

Predictor Variable

In correlational studies, the independent variable is often referred to as this.

Deception

In research is justified when the potential benefits outweigh the risks and there is no alternative way to achieve the desired results. When using deception, it is essential for the researcher to explain the nature and purpose of the deception immediately after the study concludes.

Research Methods

Involves the processes researchers employ to collect data and the procedures used to analyze the data to reach conclusions. The most reliable and valid data-gathering and data-analysis methods are fit to the research hypothesis or research question.

A portfolio

Is a collection of products that can be evaluated to determine the quality of an individual's performance.

Hypothesis

Is a tentative explanation for a phenomenon and is used as the basis for further investigation. Considered a prediction, an educated guess, or a statement of what specific outcome the researcher expects. In this step, we must consider and establish the variables (factors, characteristics, and conditions) that are under investigation. Developing the hypothesis(es) is critical to designing the appropriate procedures to collect data and extract meaning from the results.

Variable

Is any behavior or trait that varies under different conditions. (Eating, strength, satisfaction, personality traits, height, weight). Measurable variables or the effects of one variable on another because it helps them to build theory about how we behave or to understand who we are.

Data Analysis section - Includes Descriptive statistics, Inferential statistics, Parametric Tests, Nonparametric Tests

Is the process of analyzing collected data and describing any statistical procedures conducted. When data from an entire population are analyzed, the descriptive tools (mean, and standard deviation of the population) are referred to as parameters. When data from a sample are analyzed, the descriptive tools (mean, and standard deviation of the sample) are referred to as statistics. Remember populations yield parameters; samples yield statistics.

Informed Consent

Means that a participant must have full knowledge of the potential risks, benefits, procedures, and purpose of the study. Free to withdraw participation at any time.

Nonresponse bias

Occurs when participants do not complete or return the instrument. A second concern is that certain level of literacy is required to understand and complete the instruments. It is a good idea to keep the questions simple and closed-ended, giving the opportunity for participants to expand upon a response if given.

Operational definitions

Outline the precise steps required to measure a variable accuracy. Guidelines for evaluating operational definitions: 1. Is each definition adequate? 2. Does each definition provide a complete description of the important dimensions of each variable? 3. Are the definitions accurate? 4. Are the definitions of the variables universally agreed upon? 5. Are the definitions clear? 6. Are the terms of measurement and devices described adequately?

Checklist items

Participants can mark the response choice that applies. Asked to check the "best" choice or all that apply. The more checked the more complicated the data analysis and interpretation become.

Null Hypothesis (H0)

Predicting no differences (or relationships) will be observed between or among groups.

Alternative hypothesis (H1)

Predicting the differences (or relationships) will be observed between or among groups.

Directional Hypothesis

Predicts not only the difference (or relationship) but the direction as well. "Participants with high counselor-client rapport will experience higher counseling outcomes than participants with low counselor-client rapport."

Nondirectional Hypothesis

Predicts that a difference will occur without indicating which group will perform higher. Will not predict whether the relationship between variables in an associational hypothesis will be positive or negative. "The counseling outcomes among participants with different levels of therapist-client rapport will vary."

Open-ended items

Provide a question or incomplete sentence stem that cannot be answered yes or no or with a single-word response. Qualitative evaluation studies use this because they yield descriptives and narratives.

Binary option items

Provide only tow, mutually exclusive answer choices for each question or prompt (e.g. yes/no, agree/disagree, true/false, satisfied/dissatisfied). Simplifies statistical analysis and interpretation but restricts participant responses to an "all or none" dimension. Attitudes or preferences are rarely so easily categorized, leading to inherent response error owing to the lack of response sensitivity.

Results Section

Provides a detailed summary of the collected data after statistical analysis. Present the resulting data analysis in a meaningful manner (tables, correlation matrices, and graphs). All figures should be constructed according to the guidelines of the American Psychological Association (APA). No interpretation is done at this point in a manuscript.

Instrumentation Section - Includes Apparatus

Refers to any scales, tests, or assessment instruments that are used in the collection of data. A detailed rationale for use of the instrument(s) should be provided in any published research study. Provide a rationale for their choice of instruments in this section by reporting information related to normative samples, reliability and validity data, and overviews of scales and subclass.

Procedures Section

Refers to the detailed description of the process of the study (what, when, where, how, and with whom) and a description of the materials used in the study. The general design or methodological process the researcher will use should be discussed here. Sometimes the procedures are very specific, even to the point the researcher gives verbatim directions. This helps to standardize the procedures so that any other researcher can replicate the study.

Sample

Refers to the individuals who will participate in a study. Through a systematic procedure, the researcher can draw a representative sample of people who have depression. This is the subset of the target population to be studied. If the sample was selected properly and appropriate safeguards used, the sample results should generalize to the population and provide external validity.

Quantitative Methods

Rely on mathematical calculations to characterize the data collected to address the research question. Seeks objectively rather than subjectively. Results are meant to be generalized from the sample being studies to the population it represents. Mainly objectivity by using instruments that are theory based and require the percipient to reply in a standardized format. Seek to generalize from the sample to the population, they are trying to confirm an effect or relationship.

Qualitative Methods

Rely on words, or narrative description, rather than numbers and calculations, to characterize the data collected to address the research question. Do not usually derive numerical data, instead relying on the participant verbalizations to try to include the subjectivity that is represented by feelings, emotions, and attitudes. Accomplished by asking responses to the data-gathering instruments in their own words rather than on an instrument where the responses were designed by the researcher. Seeks to understand how people make meaning of such effects or relationships among events. More expiatory.

Ranking items

Require participants to rank order a number of provided choices. Participants do not always rank all of the given choices creating "missing data" Greatest limitation: require use of nonparametric or descriptive statistics, not the more powerful parametric statistical techniques (e.g. t-test and pearson r).

Basic Research

Research that seeks to build theory.

Sampling Methods Section - Includes Sample and population

Researchers must carefully describe the methods and procedures that will be used to collect the sample because that will affect the degree to which inferences can be made about the population. Convenience sampling, simple random sampling, stratified random sampling, cluster sampling, purposeful sampling, snowball sampling, and multistage sampling.

Implications for Practice and Future Research Section

Researchers should address in a report or manuscript is the implications for practice and future research. Most consumers of research want to know how the results apply to them (how can they use it in their counseling session). Also because most studies are based on previous efforts to some degree, researchers often address the degree to which the findings of their study add or fail to add to the current problem or issue under investigation. Can range up to several pages of text and is the part of the article that practitioners find most helpful. Be meticulous in this section. Researchers often suggest the next research steps needed to extend upon the findings of the current and previous studies.

correlation statistics

Reveal how strongly two sets of scores are related. Facets of personality or other personal characteristics or behaviors often appear in concert (i.e. are related). Knowing how depressed someone is can help a professional counselor relate other behaviors or characteristics, such as likelihood of suicidal ideation or substance abuse.

Nominal scale

Simply classify a client characteristics into non ordered, exhaustive, and mutually exclusive categories. Merely provide labels to describe a participant characteristic (Sex, marital status, U.S Geographical area of residence). no magnitudes, equivalent intervals, or absolute zero point, few mathematical operations can be preformed on the results. Only mode (most frequent occurring score) has some relevance.

Applied Research

Takes basic research one step further into the world of practice. One sample is assigned randomly to the experimental peer counseling setting, and the other is assigned to a control setting of counseling that is not drawn from social learning theory. The study is the application of a sound theory that is now being used to change people's behavior, rather than only to test the power of the theory to explain the behavior.

Validity

The degree to which a study's conclusions are consistent with the data used. Scores derived from the instruments (the questions) may be reliable, but how the researchers interprets the findings affects validity.

Reliability

The estimate of how consistently the scores from instruments used to gather the data performed or how closely they approximated the "true" score.

Debriefing

The explanation of the study is called this. Researcher should use this time to ensure that the participant has not sustained any physical or psychological harm as a result of the study.

Population

The larger group of individuals whom the results of the study (based on the sample) are to be generalized. Determine how widely their findings apply by identifying who they would like to study (e.g. adults with depression)

Action Research

The researcher is often a practitioner who has an immediate problem he or she would like to understand better. The ultimate goal is to create programmatic interventions

Conceptual definitions

Those found in the dictionary

Dependent Variable

Those that "depend on" the independent variable for their response. The counseling session is the independent variable, and the ability to self-regulate is the dependent variable because the nature of the client's needs require the counselor to increase the client's self-regulatory practices through the sessions. These are measured or observed, not manipulated.

Histograms

Used to visually display ordinal, interval, or ratio data and thus are more useful when summarizing test scores and most other variables used in research. Usually display the measured scores along the horizontal axis and frequencies along a vertical axis. For grouped frequency distribution, each interval of scores is positioned on the horizontal axis with the inclusive values indicated by two vertical lines at the extreme limits for each value. The height of the graph is proportional to the frequency of each score.

Quantitative Variable

Variables that can be represented on a numerical scale and are not exclusively categorical- example, distractibility, height, or place finished in a race. Generally, ordinal, interval, and ratio scales measures are quantitative variables.

Categorical Variable

Variables whose response choices can be categorized - sex (male, female), candidates for political office, geographical regions of the United States. Similar to nominal scale variables-that name but do not order or quantify the choices. "Quantitative Variables"

Critical Theory (Postmodern concept)

accepts the tenet that reality is socially constructed. Acknowledge that an individual's reality is embedded in power relations within social interactions. Multifaceted and action oriented. Common goal of using research as a criticism of cultural ideology as a way to eliminate oppression. Seeks to examine how social power is reproduced in various group settings. Believe that research values or bias should influence research process and outcome, thus inviting social activism and contributing to social justice.

Inferential statistics

allows researchers to study a sample of participants using scientific methodologies to reach conclusions that may generalize to larger populations. Help counselors predict human behaviors and reach likely conclusions (within a reasonable band of error). Researchers can help determine the primary causes of conflict among couples and interventions most likely to facilitate resolution of these conflicts and promote good couples communication.

Standard error of mean

allows the researchers to establish a confidence interval around the mean within which likely percentages of randomly derived sample means are likely to fall. The formula for computing the standard error of the mean is: Where ____is the sample variance. The standard error of the mean is default statistic provided in SPSS output, as is often presented in conjunction with the mean in empirical studies. To construct a confidence interval where t equals the desired level of confidence (LOC). Confidence intervals are generally as _________, which is the 68 percent of LOC, but they are sometimes reported at the 95 or 99 percent LOC, depending on the degree of certainty required by the researcher.

Group frequency distribution

are similar to the simple frequency distribution, except that instead of reporting every score obtained, the scores are grouped into convenient class intervals (i.e. a lot of scores are collapsed into fewer groupings of scores). Important when using this, one loses some precision while gaining some advantages in understanding the visual summary.

Continuous scales

are used to measure variables that, theoretically, can be divided to provide a more precise measurement of some psychological characteristic (Likert-type, frequency estimates)

Ordinal scale

display named categories, but these categorical choices also fell in an order sequence. Allow individuals to b ranked according to magnitude or some degree of quality or frequency. (Ex. behavioral, clinical, and personality scales frequently employ ordinal scales)

Interval scale

have ordered response categories, and the magnitude between each category choice is of an equal interval. Do not have an absolute zero point, so no absolute comparisons can be made (Ex. thermometers and many standardized intelligence and achievement tests)

Autonomy

having the ability to make the own well-informed decisions.

stem-and-leaf plots

histograms turned on their sides. The frequency of a given range of scores is represented by the length of each row. A decided advantage is that the actual numerical values provided can be used to reconstruct the scores comprising the database. The column at the far left is referred to as the stem (i.e., a class interval) and the lengths of the scores projecting to the right are referred to as the leaves. It accurately identifies the raw scores comprising the distribution. Use following steps: 1. Enter the stem digits in the left column vertically in order fro the lowest to the highest or highest to the lowest. 2. Either draw a vertical line to the right of the stem column or leave a space 3. Enter each score in the row corresponding to the appropriate stem digit, forming the leaves. 4. Order the digits within each leaf from the lowest to the highest moving from left to right.

Performance indicators

include any data that can be generated to represent how a participant performs on a given task or receives how he or she performed a given task (ex. GPA or classroom grade is a performance indicator. Daily work behaviors (e.g., attendance, times out of seat without permission, or number of times washing one's hands) and attitudes (e.g. social self-efficacy and attitude toward school)

Curve smoothing

is a common procedure but should be undertaken carefully because it involves a number of interpolations (e.g., guessing within the bounds of what is known).

Frequency polygon

is also used to visually display continuous variables. A polygon is a many-sided figure. It is basically the same as a histogram, except instead of using bars rising from the horizontal baseline, a single point above the midpoint of the interval is provide. This point represents both the score value (i.e.,class interval) and the frequency of occurrence for each value. Once these points have been charted, straight lines are drawn from point to point until all the points are connected. A straight line drawn from the farthest point at each end of the graph down to the baseline to indicate where the scores end.

Median (Mdn)

is defined as the middlemost score, the value below and above which 50 percent of the cases fall. Also becomes the 50 percentile rank. To determine: first line up all scores in order from lowest to highest, then determine the number of scores, and finally locate the middle score in the rank order. If you have an odd number of scores, the median is easier to find because you divide the number of scores by 2 and round up. In 11 scores, the median score is the sixth score from either end (11 divided by 2= 5.5) If there are even number scores: Divid the number of scores by 2 (20 divided by 2 = 10 means that the 10th score from each end is the median. But there are two 10th scores, depending on which end you count in from. This means there are two middlemost scores. Now if both scores are the same, the answer is simple because the median is equal to that number. if the two middlemost numbers are different, however, you must average them to get the media score. If two middlemost scores were 12 and 13, so the median score is 12.5 (12 + 13 = 25; 25 divided by 2 = 12.5). Commonly used when it is suspected that outliners (i.e., extreme scores) may "skew" a distribution. Also common in ordinal scales scores and interval or ratio data that are skewed or truncated - occurs when scores are cut off a certain point, even though some individuals may have either higher or lower scores.

Ratio scale

is ordered, equal interval, and has an absolute zero point; thus, any mathematical operation can be applied to ratio zero. Just because a scale has a zero point does not make that point an "absolute zero point." Absolute zero means the total absence of quantity of the characteristics being measured, such as being totally out of water or gas)

Mean

is the arithmetic average of a set of scores. To obtain the man, simply add all of the scores in a distribution and divide by the number of scores: where M is the mean of the test scores,_____means to sum all of the scores in the test distribution (X), and n is the number of scores in the distribution. When computing the mean from a frequency distribution, the formula becomes: where__ is the frequency of a given score on a test (X) The mean is most commonly used measure of central tendency with interval and ratio scale scores that are fairly normally distributed, and because the mean is directly applicable to advanced statistical procedures. The median and mode are not. As the number of scores in distribution increase, so does the stabile of the mean. Unfortunately the mean is affected by extreme scores (outliners) and has limited skewed distributions. Thus, one or several unusually high or low scores can substantially change or distort the mean and reduce its usefulness as a measure of central tendency.

Mode

is the most frequently occurring score in a set of scores. It is also the only measure of central tendency that can be used with nominal data. The highest point on a curve, histogram, or frequency polygon. If two scores simultaneously occur most frequently, the distribution is called bimodal. If one score of 14 was eliminated from the original list, the distribution would be multimodal because the scores 8, 10, 12, 13, 14, and 18 would each occur twice. By this way, if every score in distribution occurs only once, it is referred to as rectangular distribution and has no mode. The mode is rarely reported as a measure of central tendency but may be of interest to know what score the most examinees obtained.

Discrete scale

leave little to no room for disagreement. Male and female are only two categories possible on the variable sex, making it a discrete scale.

Simple frequency distribution

lists every score and the number of times it occurs, generally from the highest score to lowest score.

Bar graphs

look like histograms, but the bars on a bar graph never touch each other. A method for visually displaying discrete variables (e.g., sex or religious affiliation). Basically, any nominal variable can be appropriately displayed on a bar graph. Usually, places the categories of the nominal scale on the horizontal axis (called the y-axis or abscissa) and the the frequency of occurrence on the vertical axis or ordinate).

Standardized tests

measure achievement, intelligence, anxiety, substance use, hyperactivity, career interests, andy many other behaviors.

Beneficence

promote the welfare of those involved

Forced-choice items

quite popular in testing. Most questions are some permutation of multiple-choice question format. Forced to choose a right or wrong answer. Easy to analyze

Sensory experience

recognized as one way we come to know something or to treat it as a portion of knowledge up0on which we rely or act.

Ontology

refers to the nature of reality; specifically, it involves an understanding of the things that constitute the world

Methodology

refers to the processes and procedures of inquiry.

Axiology

refers to the role of values in research, including the idea of researcher bias and reflexivity

Epistemology

refers to the study of how knowledge is acquired in the context of the researcher-informant relationship

Rhetoric

refers to the way by which data analyses are presented. Should results be numerical or narrative?

Inductive reasoning

researcher begins by observing real and practical data that are evident in the environment to better understand the data or even develop a theory to better explain the observations of the data.

Deductive reasoning

researcher begins with an established, or at let a tentative, theory and through an in-depth study of how the theory purportedly operates, develops and tests hypothesis of how certain facts of the theory will operate or relate to observable variables.

Measures of variability

tell what the distribution looks like as scores away from the middle are explored. Two distributions may have identical means or medians but very different shapes. Describes how much variation is evident in a distribution of scores. In group counseling, a professional counselor may need to prepare for sessions quite differently if all clients are female (low variability) versus male and female (more variability). The concept of variability is essential to an understanding of data. It gives helpful indications of what scores mean and what to expect fro differences between scores. Frequently report range, interquartile range, semi-interquartile range, average deviation, variance, and standard deviation.

Semantic differential scale items

use bipolar adjectives comprising the activity pairs (e.g. active-passive or fast-slow), evaluative (e.g. good-bad or dirty-clean), or potency pairs (e.g. large-small or hot-cold). In quantitative research, evaluative pairs are most common.