Accuracy: The number or percentage of correct responses by a participant (e.g. on a lexical decision task) Amazon Mechanical Turk: Online platform where researchers can post short experiments and tasks for workers (participants) to complete Answer:The response that is made to a stimulus item during an experiment. Different experimental methods involve different types of answers. For example, in a word-by-word self-paced reading experiment, a written sentence is presented on a computer screen one word a time. Participants read each word and then press a button to see the next word in the sentence. In this case, the *answer* is a simple button press indicating that the participant has read the stimulus word on screen. Other experimental methods involve more complex answer (or response) formats. For example, in a categorization task, participants might see/hear a stimulus item and then respond by pressing either button A, B or C to indicate that the stimulus belongs to Category X, Y or Z, respectively. Answer Response Time:The amount of time it takes to respond to a stimulus item during an experiment. Response times are often measured as the total number of milliseconds from when a target stimulus item was presented (i.e., stimulus onset) to when a valid response/answer was made. (For discussion of valid responses, see answer above)
Baseline condition: Condition that measures the outcome variable given some assumed basline value of the predictor variable that is compared to the experimentally manipulated value of the predictor variable. Between-participants design: Participants are divided into different groups (e.g. Control group, Experimental Group). Groups are compared to other groups in the analysis. Bi-syllabic A word composed of exactly two syllables. E.g., 'ego' and 'toothpaste' are both bi-syllabic words.
Categorical Variable: Variable that can only take on a fixed set of values. For instance, color (red, blue, green) is a categorical variable. Compare to Continuous variable. Clauses:A group of words that contains a subject and predicate. A clause can form a complete sentence, or part of a sentence. For instance, in the sentence, "Mary saw the man", the clause is the entire sentence. However, "Mary saw the man, who was wearing the red hat" contains two clauses (underlined). Clauses that can stand alone as a complete sentence on their own ("Mary saw the man") are known as independent/main clauses. Clauses that cannot stand alone ("who was wearing the red hat") are dependent/subordinate clauses. Comma Separated Value Format (CSV format): Plain text document format in which columns of data are separated by a comma (or other delimiter, such as a tab). Very computer-friendly. You can create one of these files by saving as csv in Excel. Comprehension Question: Comprehension questions provide a measure of whether participants are paying attention during an experiment (i.e. in self-paced reading, ensures that they are reading the sentence, as opposed to just clicking to get through it as quickly as possible). Comprehension questions can be asked after every sentence, block of sentences, or even at the end of the experiment (depending on the type of experiment). Confidence Interval: Measure of how well the sample statistic (e.g. the mean heights of a group of randomly selected students) approximates the actual population statistic (e.g. the actual mean heights of students if we had access to the entire population). Provides a range of values that the statistic should fall in between x% of time, depending on the confidence level. E.g. A confidence level of 95% means that if we sample from the same population numerous times, the true value of the statistic we are interested in should fall within the confidence interval in ~95% of cases. Confound: Something in the experiment that a) affects the results (i.e., is correlated with the outcome variable) and b) is correlated with the predictor variables. E.g. if we want to examine the effect word length has on reading time (e.g., two vs three syllable nouns), we should make sure to control for word frequency, since a) frequency strongly affects reading times (Rayner et al. 1989) and b) longer words tend to be less frequent than shorter words (Zipf 1949). Continuous Variable:Variable that can take on any value between two values. E.g. height is a continuous variable because people can be any height within a certain range. Compare to discrete variable. Control group: The group of participants that serves as a point of comparison for the experimental group. For example, in a clinical trial to test the efficacy of a new drug, the experimental group would receive the drug while the control group might receive a sugar pill instead. Control variable: Factors that are held constant in the experiment. For example, if we want to find out how sunlight affects plant growth, we would want to make sure all the plants receive the same amount of water and fertilizer, so that these factors will not affect our results. Corpus: A collection of written text Critical Item: Item in the experiment that you are interested in (contrast with Filler Item).
Critical Region: In self-paced reading, the critical region of the sentence refers to part of the sentence where your experimental manipulation occurs. For instance, say you are interested in how reading times are affected by semantic predictability. Your stimuli might include sentences such as "Florian ate the chocolate yesterday" (high predictability) vs "Florian ate the table yesterday" (low predictability). The critical region occurs when chocolate/table is read.
Dependent Variable: The variable that responds to the change made in the independent variable; this is the quantity or value that is being measured. For example, to test how sunlight affects plant growth, a researcher might vary the amount of sunlight (independent variable) different plants receive. The growth of the plants is the dependent variable because it depends on the independent variable.
Excel Format (XLSX): Document format created by Microsoft Excel. Exclusions: Data that won't be included in the final analysis. E.g. we may want to exclude trials with unreasonable reading times Eyetracker: Technology used to measure the position of where the participant is looking in real-time.
Factoral Design: An experiment with two or more Independent Variables. E.g. in a 2x2 factorial design, there are two independent vriables and one dependent variable. Filler Item: Items in the list that are used to make the participant unaware of what the experimenter is testing. E.g. If we are interested in reaction times to Sentence Structure A, we might also want to include some stimuli that are of Sentence Structure B. Frequency: Number of times a word occurs in the lexicon, usually as measured within some corpus.
Independent Variable: The part of the experiment that is changed by the researcher. For example, to test how sunlight affects plant growth, a researcher might vary the amount of sunlight (independent variable) different plants receive. The growth of the plants is the dependent variable because it depends on the independent variable. Iteraction: When two or more independent variables affect the dependent variable in a way that is non-additive. E.g.
Lexical Decision Task: Task in which a participant sees or hears a word, and has to determine whether it is a real word or a fake word (usually as quickly as possible). List: Order(s) in which the items will be presented in the experiment
Neighborhood Density: A measure of how many words sound similar to the word in question. E.g. "cat" and "mat" are phonological neighbors. Words with many such neighbors have a high neighborhood density.
Outcome Variable: In (behavioral) experimental design an outcome variable is a response that is measured and (hopefully) changes in response to experimentally controlled differences in a predictor or independent variable. Also known as a dependent variable. Outlier An observation point that is distant from other observations.
Participants: The people who participate in a study (sometimes called subjects). Pilot Experiment: Small scale experiment used to test and improve the experimental paradigm before the full scale version is run. Population: The complete group of items we are interested in. (Compare to sample). Power Analysis: Statistical analysis that lets us determine how big of a sample size we need to detect an effect with a certain confidence. Predictor Variable: In experimental design a predictor variable is one that an design manipulates to test for a causal link between the predictor variable and an outcome variable. Also known as an independent variable). Priming: Effect in which exposure to one stimulus affects how quickly a subsequent one is processed.
R: Statistical programming language Reaction Time (RT): How quickly the participant is able to give a response. Sentence Region: A portion of a sentence, typically where reading times are predicted to differ (or not differ) across experimental conditions. Relative Frequency: How often something happens divided by all the outcomes. Often expressed as a proportion or percentage. Replicate: To repeat an experiment in the same or similar conditions. Typically to test the reliability of an effect. Residual: After fitting a model (e.g. linear regression) to the data, the residual of each observed value is equal to the difference between the observed value and what the model predicts at that value. Plotting these residuals can help you determine whether the particular type of model was a good fit for the data.
Sample: A collection of data drawn from the population. E.g. if we want to study language development in two-year-olds, we will draw a sample from the entire population of two-year olds. (It wouldn't be possible to have access to the entire population!) Self-paced Reading (SPR): Experimental paradigm in which words of a sentence are displayed one by one in response to the participant's actions (e.g. pressing a space bar on the computer). Spill over: Often in self-paced reading studies differences in reading times can happen both on the sentence regions that are experimentally manipulated as well as several words later. For example, in a garden path sentence, reading times can slow down both at the disambiguating word and a few words after. Stimulus: What the participant is exposed to that causes a change in the dependent variable Successful participants: A participants who is included in an analysis of an experiment. I.e. completed the experimental task and was not excluded based on a prioriexclusion criteria. Syntactic class: E.g. nouns, verbs, adjectives
Word-by-word Reading: A specific type of self-paced reading task. In a word-by-word reading task, each word in a sentence is presented one at a time. Participants read each word and then make a response/answer to see the next word in the sentence.