Statistical Analysis in Psychology

Statistical Analysis in Psychology FRQ Review Last Lesson: Explain how Mrs. Schmidlap might study her research question: Independent Var Operational Def of DV Confounding Var Ethical Consideration Definition An independent variable is something that an experimenter does. Definition An operational definition is an attempt to make an abstract concept such as learning, observable and measurable. Definition A confounding variable is anything that might affect the dependent variable that is not the independent variable. Definition Ethical considerations are issues that might harm the participants and are guidelines that help researchers conduct experiments. Application Mrs. Schmidlap’s independent variable is whether the students watch educational videos. Application Mrs. Schmidlap’s needs to operationalize the dependent variable of learning so that it can be measured. Application In Mrs. Schmidlap’s experiment, one confounding variable is whether students read a book in addition to watching the video. Application In Mrs. Schmidlap’s experiment, she needs to be careful to make sure that she keeps student grades confidential and not publish them in her study. Synonyms Mrs. Schmidlap needs to operationally and precisely define “Watching videos” so that it can be measurable. She should say students must watch at least 30 min. Synonyms However, by operationalizing and simplifying, she might be guilty of reducing learning to something that it is not. For instance, students can sometimes earn 80% without having learned anything from HW. Synonyms Another confounding variable might be student motivation to learn. If students have the video playing, but do not pay attention then their lack of attention might be a confounding variable. Statistics in Research What are Statistics? Two Types: Inferential Statistics Data exploration techniques-methods designed to help us make sense of large amounts of data. Inferential statistics Descriptive Statistics. Purpose: Can knowledge be inferred from a small sample to a large population How do we show our Population? Small back to large Random sampling doesn’t always lead to a representative sample - why? How do we denote our Sample? The group you’re working with cannot symbolize the entire world, it cannot be generalized FRQ HINT: Statistical Significance (p.60) Define: The probability of some result from a statistical test occurring by chance What does p represent? Chance,error, and confounding variable. What does p<0.5 mean in psychology? How does the size of p relate to the significance of our results? Explain the negative correlation. There is a 95% chance that the study’s results are not due to randomness, bias or confounding variable. -A P-value of .01 means that there is only a 1% chance that the results are due to chance or error -There can never be a p-value of 0 because nothing can be proven 100%, including this sentence. Practice 1: Pause Video at 15:18 to Answer Before His Slide Pops Up! Mr. Reed’s results yield a p-value of .07 (p = 0.07) Is the difference significant? Level of Uncertainty = 7% Level of Certainty? 93% What Conclusion Can He Draw? Null Hypothesis Null Hypothesis Why? Null hypothesis predicts there will not be significant relationship Goal: 1.Scientists are skeptical 2. The assume nothing until they see the numbers The goal is to disprove the null hypothesis which, in inferential statistics, predicts the results cannot be applied to a larger population Summary Chart: Can knowledge be inferred from a small sample to a larger population? Inferential Error Some examples: (ignoring statistics & drawing improper conclusions) Define: Anything that allows us or encourages us to link a small group to a larger group where a link shouldn’t exist. Heuristics: Cognitive Biases: Outgroup Homogeneity: A thinking shortcut The process of thinking illogically “They are all like that” Check for Understanding: Stop video and try to answer before he does! Why inferential statistics? Descriptive Statistics Measures of Central Tendency That the small sample will apply to everyone else. Used to describe the population or sample that is being studied Mean “The average” all of scores summed and Median The score that appears exactly in the middle of Mode Most frequently occurring score divided How do outliers skew your data? Which measure is most descriptive of set when data are skewed? the whole set The median and the mode are relatively unaffected by the presence of a few extreme scores (outlier) but one extreme score can throw off the mean. Skewed Positively Graphic Representations: Be mindful you can create simple representations of provided data as this is a specific skill set in the new curriculum. Frequency Distribution Bar Chart Histogram X = independent variable X (IV) = Categorical data X (IV) = ordinal Y = Dependent variable Y (DV) =Nominal Y (DV) =Numerical Measures of Variability (Dispersion) Standard Deviation (p58) Relation b/t range & sd? Define: How far the scores differ from the mean Low V -all scores are close together High V - Scores are spread out Range - If the highest was 96 and the lowest was 54, the range=42 A quantity calculated to indicate the extent of deviation for a group as a whole. Z-scores Asses a scores distance from the mean Used for: it helps researchers compare scores from different studies with different standard deviations. Clarence? Scored a 72 on a test with a mean of 80 and a sd-8 Z score=-1 Suzie? 84 on that same test STOP! Practice Time: Jane 40 on a test with a mean of 38 Bil 600 on a test with a mean of 700 77 Daryl 50 on her test with a mean of 50 AP Exam Practice: Researchers conducted a study in which 10 participants went to a laboratory around dinner time. The researcher assigned students to one of two groups by flipping a coin. Half of the participants were assigned to a “large bowl” group, and half were assigned to a “small bowl” group. By luck of the draw, 4 men and 1 woman were in the “large bowl” group, and 4 women and 1 man were in the “small bowl” group. Participants were told they could take as much pasta as they wanted from the serving bowl in the middle of the table, which was kept at nearly a full level. Participants were encouraged to go back for as many helpings of pasta as they wanted. In order to determine how much food each participant actually ate, the researchers measured the weight of the food that each participant took, as well as the amount that the participants left in the bowls. The amount of pasta (in grams) each participant ate is shown in the table below: Large Bowl 424 258 376 387 445 Small Bowl 196 274 148 202 250 Mean: 378 Mean: 214 Std. Dev.: 72.6 Std. Dev.: 49.3 Part A: Determine the median of the “large bowl” group. Which group has a smaller range? Explain how the smaller range related to the standard deviation. Convert this frequency distribution table into a chart. (Correctly label the x and y axis) Exit Ticket: I Feel Confident/Better About: I understand how to differentiate the types. But I Still Have Questions Surrounding: I still am I little confused on standard deviation but I will watch some videos to help.