PSYCH 312 1st Edition Exam 2 Study Guide Lectures #6-101) Define reliability and validity. Can a measure be reliable but invalid? Can a measure be valid butunreliable? - Reliable: consistent, produces or shows the same effect/result repeatedly o Can be reliable without being valid- Valid: must reflect the variable you say it measures.o Cannot be valid without also being reliable 2) How do scales of measurement relate to operational definitions? - Scaling defines the rules we use to transform our DV(s) into numbers o Determines how we display, analyze, and interpret the data3) Be familiar with the four types of scales of measurement covered in class. Be able to define each and distinguish between them. - Nominal: numbers represent categories or labels only o Using a code for sex (female=1; male=2)- Ordinal: provides categories & ranking o Ranking images in terms of disgust (1=most, 2=moderate, 3=least)- Interval: values related by a single underlying quantitative dimension with equal intervals between the scale values (can be positive, zero, and negative BUT NO absolute zero)o Fahrenheit temperature (0 doesn’t represent absence of no heat)- Ratio: everything interval data has with the difference being there is an absolute zero o Food consumption, reaction time, accuracyThese notes represent a detailed interpretation of the professor’s lecture. GradeBuddy is best used as a supplement to your own notes, not as a substitute.4) What limitations are associated with each type of scaling? That is to say, what can you do with each type of data? What can you not do? What are the advantages/disadvantages of each? - Nominal: quantitative NOT qualitative o Advantages: can only calculate frequency or percentage in each categoryo Disadvantages: limits how we can analyze our data (cannot add, subtract, multiply, or divide)- Ordinal:o Advantages: numbers reflect some degree of quantitative difference o Disadvantages: differences between values are not always equal- Interval:o Advantages: can have positive, zero, and negative values; can add, multiply, subtract, divideo Disadvantages: no absolute or true zero point representing total lack or absence of value; can’t make claims based on relative magnitudes- Ratio:o Advantages: can add, multiply subtract, divide; can make relationship based on ratioso Disadvantages: zero is lowest value (can’t have negative numbers)5) How do you know what type of measurement is most appropriate for the variables you have chosen to study? What questions should you consider? - Consider your hypothesis - Consider operational definition(s) of variables o “What measure makes the most sense?”o “What instrument/device will you use?” Questionnaire? Scale? Stop watch?- Then ask..o “is it a reliable measure?”o Is it a valid measure?6) What is a frequency distribution? What does it tell us? - Frequency distribution: plot how frequently each score appears in your data- Tells us: how frequently each value of your DV appears in your data set7) Could you create either a table or a polygon if given a set of data? - Yes 8) Be familiar with the three descriptive statistics discussed in class: mean, median & mode. Defineeach. Could you calculate/identify each if given a data set? - Mean: the arithmetic average- Median: middle score (the score that cuts the distribution into two equal halves)- Mode: the most frequent score9) How do the mean and median differ from one another in terms of how they are impacted by extreme scores in the data set? What implications does this have when we are trying to choose the most “representative” value for our data set? - Extreme scores:o Mean: change 1 scorechanges the mean (extreme scores have a greater effecton the meano Median: not sensitive to every individual score in the distribution especially extreme scores- Better to use median vs mean for representative when a set of scores is heavily skewedo Median not affected by extreme scores; mean is10) How does the relationship between the mean, median and mode vary as a function of the shapeof the distribution of the data set (i.e., normal, positively skewed, negatively skewed)? - Positively skewedo Mean>median>mode (extreme scores on the right of continuum pull the mean to the right)- Negatively skewedo Mean<median<mode (extreme scores on the left of the continuum pull the mean to the left)- Normal distributiono Mean=median=mode11) What measures of central tendency are appropriate for nominal data? What about for ordinal data? - Nominal data: only mode can be used- Ordinal data: median & mode can be used12) Could you construct a bar graph or line graph if given a set of hypothetical data/results? Where does the IV go on these graphs? Where does the DV go? - IV on X axis- DV on Y axis13) What is the null hypothesis (Ho)? What is the alternative hypothesis (HA)? If we found a difference between the treatment and control conditions, what would we say about the null? Alternatively, if we did not find a difference between conditions, what would we say about the hull? - Null hypothesis (Ho): the population mean is the same as the sample mean - Alternative hypothesis (HA): the hypothesis that sample observations are influenced by non-random cause- No treatment effect: reject the null- Treatment effect: fail to reject the null14) What three questions must we consider when we are trying to determine whether any difference between the control and treatment conditions represents a true effect of the IV on the DV?- Could the results be due to chance alone?- Could the results be due to other factors?- What support exists to indicate that changes in the DV are directly related to changes in the IV15) What are inferential statistics? What are we trying to “infer” when using them? - Inferential statistics will tell us if it is “big enough” i.e., uses aspects of the sample to infer or estimate characteristics of the population 16) Why do we say that we never “prove” the HA? - We’re looking to reject the null by finding support for HA17) Be familiar with the relationship between the sum of squares, variance and standard deviation. Why do we use these calculations to formalize our discussion of the variability of our data? - Variance: dispersion/range of scores (not just low & high scores but how the total set of scores sits in relation to the mean)- Sum of squares: squared deviations- Standard deviation: transform variance into the same unit of