Knowledge Center

Center for Policy, Data, and Research

Curating knowledge, generating analyses, and strengthening dialogue



Average: Also known as the mean. Found by adding the numbers in a group and dividing by the number of numbers in the group.

Blind Study: A study in which the participants are unaware as to whether they are in the control group or the experimental group or the experimental group

Causation: One event is the result of the other event. For example, when you exercise the amount of calories you are burning per minute will go up, as exercising makes your body burn calories faster.

Cohort: A group of people who have shared a particular event together during a specific timespan. In most Head Start studies, the cohort is a classroom, center, or program year of children.

Comparison Groups: The groups that are being compared based on an important difference like a feature of family background or an intervention they participated in.

Control Group: A group that does not receive the intervention or curriculum being studied and which is compared to the treatment group that does receive the intervention.

Confounding Variable: An unseen variable that affects what is being studied.  For example, if someone graphed the relationship between shoe size and reading ability, they would find that larger shoe size is related to better reading ability. However, this does not mean that large feet cause better reading. In this case, there is a confounding variable that is not addressed - age.

Correlation: A statistical technique used to describe the strength and direction of the relationship between two variables. In the previous definition, shoe size and reading ability are correlated because when one happens, the other usually happens. They are related, but neither one directly causes the other.

Crossover: A person in a study who starts in either the control or experimental group but moves to the other group. For example, in the Head Start Impact Study many children assigned to the control group and told they couldn't enroll actually did enroll in Head Start, crossing over to the experimental group.

Data: Individual pieces of information that can be measured, collected and reported, and analyzed.

Degree (Strength) of a Relationship: A numerical measurement between 0 and 1 of the degree (strength) between two variables. The two specific strengths are Perfect Relationship (1.0) and No Relationship (0.00).

Demographics: A section of the population sharing the same characteristics such as age, sex, class, etc.

Dependent Variable: The variable being tested in a scientific experiment. For example, when studying a literacy intervention, researchers might measure children's reading levels as the dependent variable.

Double-Blind Study: A study in which neither the participants or experimenters know who is receiving a particular treatment.

Effect Size: A measure of the strength of the relationship between two variables.

Evaluation: The process of determining whether a social intervention produced the intended result.

Experimental Group: The group in an experiment that receives the treatment or intervention. The experimental group is compared to the control group, which has not received the intervention.

Generalize: Taking something specific and applying it broadly. It's important to understand whether a study can be generalized to your students. For example, a study that looks at children in major cities may have findings that generalize to other children in cities but not necessary to children in rural communities.

Independent Variable: The variable that is changed by the researcher. For example, in the literacy intervention the independent variable was which kind of literacy instruction each child got.

Institutional Review Board (IRB): A committee that has formally designated to approve, monitor, and review research involving humans. The main goal of IRBs is to protect human subjects from psychological and physical harm.

Intervention: Action taken to improve a situation.

Literature Review: A description of the existing research relevant to a particular topic. It gives an overview of what has been said, who the key writers are, what are the prevailing theories and hypotheses, what questions are being asked, and what methods and methodologies are appropriate and useful.

Longitudinal Study: A research study that involves repeated observations of the same people and variables over long periods of time - often many decades. An example of a longitudinal study is the Perry Preschool study.

Mean: Also known as the average. Obtained by adding the numbers in a group and dividing by the number of numbers in the group.

Median: In a data set, the number in the middle after the numbers have been ordered from least to greatest.  If two numbers are in the middle, then the median is the average of the two numbers.  

Mixed-Method Research: Integrating quantitative and qualitative research and analysis in a single study for the purpose of better understanding the research problem.

Mode: The number that occurs the most in a group of numbers or data set.

Negative Correlation: A relationship between two variables such that when one value increases, the other value decreases.

Norm-referenced Assessment: Tests developed by creating assessment items and then administering the test to a group of children that will be used as a basis of comparison. Examples of norm-referenced assessments are developmental screening tests and the Stanford Achievement Test (SAT).

Population: The whole group from which a sample is taken.

Positive Correlation: A relationship between two variables such that their values increase or decrease together.

Principal Investigator: The lead scientist on a particular research study.  

Probability: The extent to which an event is likely to occur, measured by the ratio of how many times an outcome happens out of a total number of outcomes. For example, the probability of flipping a coin Heads is 1:2 because it should happen once out of every two times.

Propensity Matching: A statistical matching technique that attempts to estimate the effect of a treatment, policy, or other intervention by examining two participants who are matched on all features except who got the intervention.

Qualitative Research: A method of research that focuses on understanding issues and understanding phenomena by using unstructured data. Some examples of unstructured data include: open-ended survey responses, audio recordings, focus groups, pictures, and in-depth interviews.

Quantitative Research: A method of research that focuses on gathering numerical data and generalizing it across groups of people or to explain a particular phenomenon.

Random Assignment: An experimental technique used to assign subjects to different treatment groups. This means that each individual has an equal chance of being assigned to either group.

Randomized Control Trial: A study design that randomly assigns participants into an experimental group or a control group. As the study is conducted, the only expected difference between the control and experimental groups in a randomized controlled trial (RCT) is the outcome being studied.

Sample: A group of people selected from the population to find out something about the population.

Standard Deviation: The average amount by which individual items in a data set differ from the mean of all the data in the set.

Standard Score: Standard Scores are used in norm-referenced assessment to compare one child’s performance on a test to performance of other children his/her age.

Statistics: The study of the collection, analysis, interpretation, presentation, and organization of data. There are different types of statistical tests, which include: ANOVA (Analysis of Variance), Chi square, Regression Analysis, and T-test.

Statistically Significant: A finding is statistically significant if the intervention caused an outcome at a higher or lower rate than could have occurred purely by chance.

Theoretical Viewpoint (or Perspective): A set of assumptions about reality that guides the researcher’s study.

Treatment Group: The group that receives the treatment or the intervention. This group is compared to the control group.

Validity: The extent to which a test measures what it claims it measures. For example, if a developmental test is given in a language that is not the child’s native tongue, it may be measuring the child’s language skills rather than his or her development skills in the second language. Therefore, this is not a valid test.   

Variable: An attribute that describes a person, place, thing, or idea. Variables can “vary” from one entity to another. They can be categorized as qualitative (categorical) or quantitative (numerical). For example, the number of words in a child's vocabulary could be a variable.

Variance: A statistical measure that tells us how measured data vary from the average (mean) value of the data set.