PSYC 104 Unit 1

Juan wishes to survey his fellow students about their eating habits. He asks on his survey "How healthy did you eat last week?" The possible responses are: Very Unhealthy, Unhealthy, Slightly Unhealthy, Slighly Healthy, Healthy, Very Healthy. Juan believes that the interval between each response is fairly equal.When recording the data, Juan creates the variable 'EatHealthy'. He records the data as follows: Eating 'Very Unhealthy' is recorded as a 1... up to eating 'Very Healthy' which is recorded as a 6. The scale of measurement for his variable 'EatHealthy' is:

"High Ordinal" Juan has his survey respondents use what is called a likeart scale, where the responses are Very Unhealthy, Unhealthy, Slightly Unhealthy, Slighly Healthy, Healthy, Very Healthy. Juan believes that the interval between each response (e.g., the interval between 'Very Unhealthy' and 'Unhealthy' is fairly similar to the interval between 'Slightly Healthy' and 'Healthy').In this case, the data would be considered 'High Ordinal.' It is ordinal in the sense that we can rank the responses. It is NOT interval, since we can't prove that the interval size is equal. So it is referred to as 'High-Ordinal'... that is to say... it is more than just ordinal. When analyzing the data, we'd treat it as scale data (since with scale data, the assumption is that there are equal intervals). Juan leaves a blank line after the question where the respondents can record the number of meals eaten in the prior week that included a fruit or vegtable. The number does not represent a category... rather it represents an amount. And eating eight meals with a fruit or vegtable is twice as many meals as only four meals with fruit and vegtables. So the 'EatHealthy' variable in this case would be ratio. Juan allowed the survey respondents to answer 'yes' or 'no' to the question: Do you eat healthy? These two possible responses ('yes' and 'no') represent two categories. The number assigned to each response is arbituary. Juan could decided to record each 'yes' as a 2 and each 'no' as a 1. The numbers (1 2) are simply labels for the yes and no categories. Since it is arbituary how we record the yes and no values (i.e., which one is recorded as a 1, and which one is recorded as a 2 doesn't matter), this is nominal data.

The frequency for the class interval is 17. The total number of values is 33. Please calculate the relative frequency for the class interval (to the hundredths place). The answer is...

0.52 = 17/33

What is the interquartile range for the following scores? 2, 5, 9, 12, 18, 22, 45

17

What is the frequency of people who responded 'Yes' to being vegetarian?

2 -Frequency refers to the number of occurrences. Count how many people responded 'Yes' to being vegetarian... that is the frequency.

What is 27.0495 rounded to the thousandths place?

27.05

The median for the following set of data is: 30, 15, 40, 85. Report the results without rounding.

35

What is 35.565 rounded to the hundredths place?

35.57

What is the interquartile range for the following scores? 8, 15, 20, 37, 45, 56, 68

41

If the variance for a group of scores equals 25, what is the standard deviation?

5

The Mode for the following set of scores is: 66, 5, 51, 66, 87

66

Consider a normal distribution with over 1,000 values. What percentage of values are within one standard deviation of the mean? This includes the values from one standard deviation below the mean to one standard deviation above the mean.

68%

Consider a normally distributed population of scores, where the mean is 50 and the standard deviation is 20. What percentage of scores are between 30 and 70?

68%

Consider a positively skewed distribution with a mean of 80 and a standard deviation of 16. What percentage of scores are between 64 and 96?

68% of the population is within one standard deviation of the mean, and that 95% is within two standard deviations of the mean. However, we are told that the distribution was Positively Skewed.... so we can't apply the Empirical Rule to help us out here. The Empirical Rule is only helpful when describing a normal distribution. For a non-normal distribution (skewed, bi-modal, flat, etc), the rule would be misleading. So, always check that the distribution is normally distributed before applying the Empirical Rule.

The median for the following set of data is: 4, 8, 1, 19, 19

8

If the variance for a group of scores equals 71, what is the standard deviation? Round to the nearest tenth.

8.4

The median for the following set of data is: 2, 9, 5, 17, 13

9

Consider a normally distributed population of scores, where the mean is 100 and the standard deviation is 16. Approximately what percent of scores will be between 84 and 116?

95%

A histogram communicates the same information as a:

A histogram communicates the same information as a grouped frequency distribution. One shows the information as a graph, and the other as a table.

A researcher hypothesizes that talking to plants enhances their growth. To test her hypothesis, she gives two groups of plants the same amount of food, water, and sunlight, but she talks to one group of plants and not the other. What is the independent variable?

Amount of talking (0 hours vs. 3 hours)

When would you use an experiment rather than a correlational study?

An experiment should be tested when a researcher wants to manipulate one of the two independent variables. Once they manipulate an independent variable, they test and observe if a difference was made between the two. Unlike correlation research which makes a prediction before the study if two variables are related. So an experiment should be used to determine if there is a cause-and-effect relationship between the variables.

When the measure of central tendency is the median, and you are concerned about outliers, which of the following is the most appropriate measure of variability?

Both the range and the standard deviation are affected by extreme values. Only the interquartile range is unaffected by extreme values (since it reports the range for the middle 50% of the values - from the 25th percentile to the 75th percentile).

Sally predicts that sleep and stress are related. She then goes an collects her data. Type of research method:_______________________

Correlational study

Dr. Mohr is interested in comparing the number of students earning an A in Psychology versus Biology. He learns that 700 of the 5,000 Psychology students earned an A. He also learns that 1,000 of the 10,000 Biology students earned an A. When reporting this finding with a table, Dr. Mohr should provide...

Dr. Mohr wants to compare the number of students earning an A in the Psychology and Biology departments. This means he can report either the frequency or percent. He would only report the frequency of A's for both departments if both departments had the same number of students. Otherwise, reporting the percent is a clearer way to make a comparison.

Susan decides to study how many hours people spend each day watching TV. She reports the number of people who watch (zero to four hours), (five to nine hours), etc. For example, based on her table, it appears that most people watch 5 to 9 hours of TV. How has she organized her data?Hours Watched Number of People 10 - 14 1 5 - 9 16 0 - 4 11

For a grouped frequency distribution, we divide the range of scores into class intervals, and list the number of values in each class interval.

What elements of a graph should you carefully check, to help reduce the likelihood of someone misleading you?

I studied this last semester in a Media and Information Literacy class. Whether unintentional or with the intent to deceive, it is common to find misleading information in the form of easy to read graphs. Sometimes the baseline is omitted, or the scale is incorrectly sized- both can skew data, as can using the wrong kind of graph. One should always analyze the graph title and descriptive labels on the axes to verify the data is not misleading.

___________________ statistics are used when generalizing from a sample to a population.

Inferential statistics involve probability, and evaluating whether to generalize a result from a sample to a population.

Juan wishes to survey his fellow students about their eating habits. He asks on his survey "How often do you eat 'fast food' (e.g., McDonalds') during the week?" The possible responses are: Never, Once a Month, Once a Week, More than Once a Week.When recording the data, Juan creates the variable 'EatHealthy'. He records the data as follows: For 'Never' he records a 4, for 'Once a Month' he records a 3, for 'Once a Week' he records a 2, and for 'More than Once a Week' he records a 1.The scale of measurement for his variable 'EatHealthy' is:

NOT- Ratio

Joey manipulates the temperature of the room, and then measures the person's body temperature (degrees Fahrenheit). The scale of measurement for the dependent variable is __________________.

NOT- Ratio Joey manipulates the temperature of the room, and then measures the person's body temperature (degrees Fahrenheit). The scale of measurement for the dependent variable is interval. Degrees Fahrenheit is interval scale because there are equal intervals between temperatures (e.g., difference between 9 and 12 is the same as the difference between 109 and 112). But 60 degrees Fahrenheit is NOT twice as hot as 30 degrees Fahrenheit (i.e., we can't use ratios like 'twice as much'). So degrees Fahrenheit qualifies to be interval ratio (because of the equal intervals), but does not qualify to be ratio, since you can not do ratios (such as twice as hot, etc) with Fahrenheit.

On Greg's survey, his dependent variable is the amount that a person eats. He asks on his survey "Do you consider yourself a person who eats (A) more than most, (B) similar to those around you, or (C) less than most." The scale of measurement for the dependent variable is _________________.

NOT-Nominal Ordinal On Greg's survey, his dependent variable is the amount that a person eats. He asks on his survey "Do you consider yourself a person who eats (A) more than most, (B) similar to those around you, or (C) less than most." The scale of measurement for the dependent variable is ordinal. A variable is considered ordinal if what is measures can be ranked... but the distance between the responses does not consist of equal intervals. In this case, we can rank the responses to Greg's question: Such that 'eating more than most' is greater than 'eating similar,' and 'eating similar' is more than 'less than most.' So we can rank the responses, but there is no guarantee that the distance between 'eating more' and 'eating similar' is the same as the distance between 'eating less' and 'eating similar'.... we just know that it is easy to rank the responses.

The use of random selection

NOT-insures that there will be no sampling error Random selection makes sure that every member of the population has an equal chance of being included in the sample. It also removes bias in selecting participants (since the researcher does not get to directly pick who will be in the study). While random selection helps to form a representative sample, this does not mean that there will not be any sampling error... the sample will not exactly match the population.

Barbara is interested in the effect of rewards on creativity. Her independent variable is whether a child is given 'first place,' 'second place,' or 'third place' for their artwork (randomly assigned), and her dependent variable is how the child then feels about their own art. In this case, the scale of measurement for the independent variable is __________________.

Ordinal Barbara is interested in the effect of rewards on creativity. Her independent variable is whether a child is given 'first place,' 'second place,' or 'third place' for their artwork (randomly assigned), and her dependent variable is how the child then feels about their own art. In this case, the scale of measurement for the independent variable is ordinal. The independent variable's scale of measurement is ordinal, because we can rank '1st,' '2nd,' and '3rd' place... but there are not necessarily equal intervals between each of those values. In a typical competition, whomever earns first place might have done much, much better than either second or third place... so the distance between 1st, 2nd, and 3rd place is not necessarily equal.

We never say ___________ when working with a sample (as the sample might not be representative)

Proved

Why is random assignment important when conducting an experiment? How is random assignment different from random selection?

Random assignment is crucial when a researcher wants to prove causal claims based on the studied treatment. Random selection is when a sample group is collected from a population, and occurs when every member of a population has a chance of being included in the sample, and random assignment is when the sample is randomly divided into a treatment or control group member.

John's independent variable is the amount of aspirin given to participants. The scale of measurement for the independent variable is _____________________.

Ratio

Sally's dependent variable is the number of times a person says 'uh huh' during a lecture. The scale of measurement for this dependent variable is ___________________.

Ratio

Which is the appropriate graph to communicate the results... regarding the number of vegetarians?

Remember that a graph must have a title, and the x- and y-axis must be labeled. Next, the variable 'Vegetarian' is nominal, as the responses are categorical (either 'yes' or 'no'). When the variable is nominal or ordinal, we use a bar graph (i.e., the bars do not touch - as we do not assume equal intervals between each listed response).

Greg wishes to learn more about how people respond to trauma. He randomly selects eighty people living in San Diego county. These eighty people form the:

Representative Sample The population represents everybody you wish to know about. A sample is a subset of the population. A representative sample means that the sample is similar to the population. One way to get a representative sample is to use random selection.

Greg reports the number of units taken by the students in has class as follows:Units Number of People 18 2 15 7 12 12 9 8 6 3 3 1 He has reported his data as (a): ____________________

Simple frequency distribution

The more representative our sample is of the population, the _______________ the sampling error.

Smaller

Imagine that we wish to run an experiment to evaluate whether consuming warm food (versus cold food) makes us feel friendlier towards others. What would be your Independent variable and your Dependent Variable? So that someone else could repeat your experiment with exactness and precision... How would you operationalize your Independent variable? How would you operationalize your Dependent variable?

The independent variable is the cold or warm food. The dependent variable is is how friendly you are towards others. You can operationalize your independent variable by giving each person in one group a cup of cold soup (e.g., 60 degrees fahrenheit) to consume and the other group one cup of warm soup (78 degrees fahrenheit) to consume. You can operationalize your dependent variable by counting the number of times each of the groups uses friendly gestures towards someone, such as saying please, thank you, excuse me, etc.

Why can we describe an entire population of several billion people, based on just a sample of 3,000 or fewer individuals?

The process of random selection allows researchers to choose a sample size that is representative of an entire population. It does this by avoiding biases and ensuring that there is an equal opportunity for every member of a population to be included in the research that is being done. Therefore, the sample size can provide an accurate representation of a much larger population.

If the standard deviation of a group of 4 scores is 18, what is the variance?

The standard deviation is equal to the square root of the variance. So working backwards, if we have the standard devation, we would have to square it to get the answer. Ignore the size of the sample.

When would you want to record your data (e.g,. 1, 4, 2, 4) as though the numbers were names? What is this scale of measurement called?

This is called the Nominal Scale of measurement and would be used when all categories are considered equal (e.g., male and female, Democrat and Republican, etc.). There is no mathematical connection between the numbers, so neither adding or subtracting would be appropriate.

What is the mean for the data set: 3, 17, 8? Round to the hundredths place.

To calculate the mean, you add up all of the value... and then divide by the total number of scores. For example, the mean of 5, 2, 8 is equal to (5 + 2 + 8) / 3 = 15 / 3 = 5.00. Note that with rounding to the hundredths place, a value such as 4.3333 would become 4.33, while a value of 4.6666 would become 4.67.

What is the variance for the scores 1, 3, 3, 5?

To calculate the variance, first determine the mean. In this case the mean is 3.Next, square the difference between the mean and each score (1-3=2; 2 squared is 4; do this for each score).Last, add up the squared difference scores, & divide by the number of scores.

With a bar graph, the bars ________________ because each bar represents _____________.

With a bar graph, the bars do not touch because each bar represents a different category.

A relatively small group selected from a population is called?

a sample

An experiment can consist of just...

an independent variable and a dependent variable

To calculate the range of the data set:

calculate the difference between the smallest and the largest score

Elaine is measuring the speed of cars driving on her street. This is an example of...

continuous data

Elaine is measuring the number of words a person correctly recalled. This is an example of...

discrete data

Since the research hypothesis is a prediction, it is written in ______________ tense.

future

The wider the normal distribution, the...

greater the variability of scores

An industrial psychologist wishes to study how room temperature affects productivity. She places 90 factory workers into a special room to find out if they can assemble computer components more rapidly when the temperature of the room is either 60 degrees or 80 degrees Fahrenheit. The dependent variable of this research is

how rapidly the components assembled

A histogram is used to display _____________ data in a graph format.

interval & ratio

The scale of measurement referred to as 'Scale' includes both the _______________ and the _______________ scales of measurement.

interval and ratio

A frequency histrogram of a distribution with a negative skew will have a tail pointing toward the

left. A normal distribution is symmetrical (both sides look the same). A positively skewed distribution has its tail pointing to the right.

If there is an outlier (regardless of whether the distribution is normal, positively skewed, or negatively skewed), we should report as our measure of central tendency the:

median

The ______________ is the middle value of any set of ranked scores.

median

When using SPSS, we would create a bar graph for __________________ data and a histogram for ___________________ data

nominal; scale

As long as the scale data is ________________, then we report the mean and standard deviation. If the scale data is ______________ or has an outlier, then we would instead report the median and the IQR.

normally distributed, skewed

In answer to John's survey question, people respond with a 1, 2, 3, or 4.1 - I eat more than 6 vegetables a day2 - I eat between 4 to 6 vegetables a day3 - I eat between 1 to 3 vegetables a day4 - I eat zero vegetables a dayThe scale of measurement for these responses is:

ordinal scale because while we can rank the responses, the interval between each of the numbers is not the same.Nominal Scale: Each number represents a different category (that can not be ranked)Ordinal Scale: Each number represents a category, and we can rank the categories'High Ordinal' Scale: Each number represents a response that can be ranked, and their is reason to assume that the distance between each category is fairly similiar (e.g., 1 - strongly disagree, 2 - disagree, 3 - slightly disagree, 4 - slightly agree, 5 - agree, 6 - strongly agree; we might assume that the distance between each response is fairly similiar)Interval Scale: Each number represents a value that falls along a number line (i.e., equal distance between each number; the distance between 1 to 2 is the same as 3 to 4)Ratio Scale: Each number represents a value that falls along a number line; and there is a true zero, such that 4 is twice as big as 2.

sigma

population standard deviation

When a person organizes a set of scores, listing the number of times each score occurs, this is called a...

simple frequency distribution

The measurement of a person's ________________ would involve the use of an interval scale of measurement.

temperature (in degrees Celsius)

An industrial psychologist wishes to study how room temperature affects productivity. She places 90 factory workers into a special room to find out if they can assemble computer components more rapidly when the temperature of the room is either 60 degrees or 80 degrees Fahrenheit. The independent variable of this research is

the temperature of the room

Jack measures a person's g.pa. and the length of their hair. If Jack were to find that length of hair was not related to g.p.a., this would be an example of a:

zero correlation