Brainscape's Knowledge GenomeTM

Browse over 1 million classes created by top students, professors, publishers, and experts.


See full index

IT 497 - Research Methodology > Test 3 > Flashcards

Test 3 Flashcards

1
Q

Define descriptive research.

A

Descriptive research is research that examines a situation as it currently is; it doesn’t introduce changes or modify anything. It’s a study of the nature of how things are.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

What is an observation study like in quantitative research?

A

In quantitative research, an observation study studies specific phenomena, and has a particular, pre-specified focus. That behavior is then quantified, usually counted or rated.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

What are some ways of ensuring that an observation study remains objective?

A

Some strategies to ensure that an observation study remains objective are to:
- Define the behavior being studied in highly precise manner, so it is easily recognizable when it occurs.
-

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

What are some ways of ensuring that an observation study remains objective?

A

Some strategies to ensure that an observation study remains objective are to:

  • Define the behavior being studied in highly precise manner, so it is easily recognizable when it occurs.
  • Divide the observation period into small segments and record if the behavior is seen during that segment.
  • Have multiple people rate the same behavior independently.
  • Train the raters to use specific criteria when counting or evaluating the behavior, and train until there’s consistency.
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

What are some ways of ensuring that an observation study remains objective?

A

Some strategies to ensure that an observation study remains objective are to:

  • Define the behavior being studied in highly precise manner, so it is easily recognizable when it occurs.
  • Divide the observation period into small segments and record if the behavior is seen during that segment.
  • Have multiple people rate the same behavior independently.
  • Train the raters to use specific criteria when counting or evaluating the behavior, and train until there’s consistency.
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

What is a correlational study?

A

A correlational study is a study that examines the extent to which differences in one characteristic/variable are related to differences in one or more other characteristics or variables (a.k.a. a correlation).

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

What is correlation?

A

Correlation is a link between two characteristics or variables that, when one increases, the other increases or decreases in a somewhat predictable pattern.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

What type of graph is best for charting correlation?

A

A scatter plot (or scattergram) is the best tool to plot the relationship between two variables (each variable being on one axis). If the two variables are correlated, they’ll appear to form an elliptical, sausage-y shape. Otherwise, they’ll be all over.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

What is a line of regression?

A

A line of regression is a line used on a scatter plot that reflects a perfect correlation between the two variables (i.e. a 1-to-1 relationship).

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

How is correlation described?

A

Correlation is described using a correlation coefficient, a number on a scale from one to negative one, with one meaning a perfect, direct relationship, and negative one meaning a perfect, inverse relationship.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

What is the biggest hazard of correlational studies?

A

CORRELATION DOES NOT EQUAL CAUSATION.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

What is the biggest hazard of correlational studies?

A

CORRELATION DOES NOT EQUAL CAUSATION.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

What is a cross-sectional study? How does it differ from a longitudinal study?

A

A cross-sectional study is a study where people from different age groups are sampled and compared. Longitudinal studies, by contrast, would simply take a group of people and follow them over the course of time.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

What are some downsides of the longitudinal study?

A

Some common downsides of a longitudinal study are that it’s very common to lose participants, and that over time, a person may become better at a measurement instrument due to practice.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

What are some downsides of the cross-sectional study?

A

Some downsides of a cross-sectional study are that members of different ages experienced different upbringings and experiences in general due to time period. We also cannot compare correlations between the age levels.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

What is a cohort-sequential study?

A

A cohort-sequential study is a study that combines the techniques of both the cross-sectional and longitudinal studies, tracking multiple age groups over a period of time.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

What is a cohort-sequential study?

A

A cohort-sequential study is a study that combines the techniques of both the cross-sectional and longitudinal studies, tracking multiple age groups over a period of time.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

What is survey research?

A

Survey research, in the areas of observational research, refers to acquiring information about one or more groups of people (on characteristics, opinions, etc) through asking questions and tabulating answers.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

What is a normative survey?

A

A normative survey, also known as a descriptive survey, is a survey whose goal is to learn about a large population by surveying a sample of that population.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

What is the biggest limitation on data gathered by a survey?

A

The biggest limitation on data gathered by a survey is that the information goes stale; that is, the things written on them may change as the participant goes on with their life.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
21
Q

What are some limitations of using surveys?

A

The biggest limitation on data gathered by a survey is that the information goes stale; that is, the things written on them may change as the participant goes on with their life. Another limitation is that survey data is self-reported, that is, the participants are telling us what they believe, which might not be true.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
22
Q

What’s a structured interview?

A

A structured interview is one where the interviewer asks a standard set of questions, and that’s it.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
23
Q

What’s a semi-structured interview?

A

A semi-structured interview is an interview where the interviewer may follow the standard questions with one or more individually tailored questions to get clarification.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
24
Q

How are the “feels” of interviews different between qualitative and quantitative studies?

A

The feel of a qualitative study tends to be more informal; a quantitative study tends to feel more formal and emotionally neutral.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
25
Q

What form of interview yields the highest response rate?

A

The face-to-face interview yields the highest response rate. Next highest to face-to-face is the phone interview.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
26
Q

What are some pros and cons of the questionnaire?

A

The questionnaire allows some measure of privacy/anonymity, and therefore may be more truthful answering questions on controversial topics. Data can be gathered at low cost from a large number of people.
The downside is there’s a low response rate, the questions may be misinterpreted, and people might not have great reading/writing skills. Not only that, the questions are limited to exactly what’s put down. No follow-up.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
27
Q

What is a Likert scale?

A

A likert scale is the common “Strongly Disagree” to “Strongly Agree” scale.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
28
Q

What is a perfect correlation?

A

A perfect relationship is an association between two variables that is either +1 or -1.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
29
Q

What is a strong correlation?

A

A strong relationship is an association between two variables that is close to +1 or -1.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
30
Q

What is a weak correlation?

A

A strong relationship is an association between two variables that is close to 0.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
31
Q

What is probability sampling?

A

Probability sampling is when a sample is taken from the overall population via random selection.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
32
Q

What is simple random sampling?

A

Simple random sampling is the most “simple” form of probability sampling, where every member of a population has an equal chance of being selected. Well-suited for small populations.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
33
Q

What is stratified random sampling?

A

Stratified random sampling is when you’ve got a population with subgroups (ethnicity, age, income), and you take a random group of people from each strata. We treat each strata as if it’s equal in size.

34
Q

What is proportional stratified sampling?

A

Proportional stratified sampling is similar to stratified random sampling, but instead of assuming that each strata is equal in size, we select our sub-sections according to the ratio between the strata.

35
Q

What is cluster sampling?

A

Cluster sampling is when a large population exists, and we split it into smaller units, which we then select a random subset from (think dividing a city into precincts).

36
Q

What is systematic sampling?

A

Systematic sampling is when you select people (or clusters) using a predetermined sequence, which itself must not originate by chance. For example, you might shuffle a deck of cards, and choose every fifth card.

37
Q

What is nonprobability sampling?

A

Nonprobability sampling is a method of sampling whereby the researcher has no way of telling or promising that each element of the population will actually be represented.

38
Q

What is convenience sampling?

A

Convenience sampling, aka accidental sampling, you’re not even trying to identify a representative subset. You just take people (or units) that are readily available (like surveying customers of a restaurant that come in).

39
Q

What is quota sampling?

A

Quota sampling is a variation of convenience sampling, but it makes an attempt to select respondents in the same proportions they’re found in the general population.

40
Q

What is purposeful sampling?

A

Purposeful sampling is sampling where people or units are chosen for a particular purpose. For example, people we decide are “typical” of one group or other.

41
Q

What is the typical return rate for a survey mailed to strangers?

A

The typical return rate for a survey mailed to strangers is about 50%.

42
Q

What is a parameter?

A

A parameter is a characteristic or quality of a population that, in theory, is constant. Its values may be variable, but it should exist. For example, a circle’s radius is a parameter.

43
Q

What is a continuous variable?

A

A continuous variable is a variable that has an infinite amount of possible values, falling along a continuum. An example would be chronological age.

44
Q

What is a discrete variable?

A

A discrete variable is a variable with a finite, and generally small number of possible values.

45
Q

What is nominal data?

A

Nominal data is data in which numbers are used only to identify different categories of entities (1=male, 2=female, etc).

46
Q

What is ordinal data?

A

Ordinal data is data in which numbers reflect an “order” or sequence. It doesn’t tell us anything about the magnitude of difference between groups, but it does tell us the degree that a group has a certain characteristic or variable that is being measured.

47
Q

What is interval data?

A

Interval data is data that reflects equal units in measurement. This data can show us the difference in degree, amount, etc, but it can also tell exactly how much difference exists between objects, in regards to the characteristics being measured. Zero isn’t a thing of significance, though. IQ is this, as an IQ of zero doesn’t equal a rock.

48
Q

What is ratio data?

A

Ratio data is data that is similar to interval data, but with the addition of a true zero point. This would be something like income level, or temperature.

49
Q

What is a normal distribution?

A

A normal distribution, also known as a bell curve, is the general pattern things seem to fit in. Normal distribution has:

  • Horizontally symmetry (one side is mirroring the other).
  • A high point at the midpoint.
  • Predictable percentages of population at any given point of the curve.
50
Q

What are the percentages of populations within one standard deviation of the middle of a normal distribution? Within two? Beyond that?

A

The population that lays within one standard deviation of the median of a normal distribution is 34.1%, on each side. Within two standard deviations, add an additional 13.6%. Past that, the final 2.3%.

51
Q

What do we call a distribution that doesn’t fit the normal distribution? Are there variations?

A

A distribution that doesn’t fit the normal distribution is called a skewed distribution. A distribution can be positively or negatively skewed. A positively skewed distribution occurs when the peak lies to the left of the midpoint; a negatively skewed distribution occurs when the peak lies to the right.

52
Q

What is kurtosis? Are there variations?

A

Kurtosis occurs when the peak of a distribution is unusually shaped. Leptokurtic curves are unusually pointy, and platykurtic curves are unusually flat.

53
Q

What are the three most widely used measures of central tendency?

A

The three most common measures of central tendency are mode, median, and mean.

54
Q

What is a point of central tendency?

A

A point of central tendency is a middle number around which the data for a specific variable revolves around.

55
Q

What is mode?

A

Mode is a measure of central tendency, and is the one that tends to occur most frequently. It measures which value in a data set occurs most frequently.

56
Q

What is median?

A

Median is a measure of central tendency, identifying the numerical center of a data set, with exactly as many scores above as below it. If there are an even number of scores, it’s the average of the two middle scores.

57
Q

What is mean?

A

Mean is a measure of a central tendency that gets the average of a set of data.

58
Q

What measure of central tendency is most commonly used in statistical analyses and research reports?

A

The mean is the measure of central tendency that’s used the most with statistical analysis and research reports. It should only be used when numbers reflect equal intervals along a particular scale (can’t get the mean of men and women, etc).

59
Q

What measure of central tendency is most commonly used with ordinal data, or data that is highly skewed in one direction or the other?

A

Median is the most appropriate measure of central tendency for when dealing with ordinal data, or skewed data (like, 1, 2, 3, 4, 8, 125).

60
Q

What are parametric statistics? How are they different from nonparametrics?

A

Parametric statistics are statistics that are based on specific assumptions about the population in question; the most common are:
- The data reflect an interval or ratio scale.
- The data fall in a normal distribution.
Nonparametric statistics are not based on assumptions, and are best suited for ordinal data.

61
Q

Why ever use parametric statistics? Why not just stick with nonparametric?

A

We use parametric statistics because nonparametric statistics are only truly useful with simple analysis.

62
Q

What is the geometric mean?

A

The geometric mean is a measure of central tendency that is best suited for dealing with reporting growth data. The geometric mean is calculated by multiplying all of the scores together, and then finding the Nth root of the product (where N equals the number of items in the data set).

63
Q

What is the simplest measure of variability?

A

The simplest measure of variability is range, which is the spread of the data from the highest value to the lowest value.

64
Q

What is interquartile range?

A

Interquartile range is a measure of variability that divides a distribution into four equal parts, with quartile 1 lying at a point where 25% of the members of the group are below it, quartile 2 divides the group into two equal parts and is identical to the median, quartile 3 is where 75% of values are below it.

65
Q

What is the most common measure of variability used in statistical procedures?

A

The most common measure of variability used in statistical procedures is standard deviation

66
Q

How is standard deviation calculated?

A

Standard deviation is calculated by taking the square of the differences of each score, subtracted by the mean of all the scores. Then we add together all the squared differences, and divide by the number of scores, and then finally find the square root of that.
Note, we square the differences first, then add them together.

67
Q

What are raw scores?

A

Raw scores are the number of correct answers or points that a person gets on a test or other measurement instrument.

68
Q

How do we provide context for raw scores?

A

We provide context for raw scores using norm-referenced scores, that is, scores that reflect where each person is positioned relative to other members of the person’s group. Percentile rank is a form of this.

69
Q

What are standard scores? What are some examples?

A

Standard scores are a form of norm-referenced score, telling us how far an individual’s performance is from the mean. A key example of a standard score is the z-score.

70
Q

What is a z-score? How is it calculated?

A

A z-score is a standard score showing how far an individual’s performance is from the mean. It’s calculated by taking an individual’s raw score, subtracting the mean of the population, and dividing by the population’s standard deviation.

71
Q

What R command is used to get the structure of an R object? What does it provide?

A

str() is the R command used to get the structure of an R object, and it returns the number of observations (rows) and variables (columns).

72
Q

What command is used to get the mean of a variable in R?

A

The mean() command is used to get the mean of a variable in R.

73
Q

What command is used to get the median of a variable in R?

A

The median() command is used to get the median of a variable in R.

74
Q

What command is used to get the mode of a variable in R?

A

Getting the mode in R isn’t as simple as just running a command. First, we must get the variable as a table using
temp Getting the mode in R isn’t as simple as just running a command. First, we must get the variable as a table using
temp = table(as.vector(nihs$SLEEP))
Which produces a table with two rows, one listing each data value, and the second listing how many times those variables occurred. The mean can then be obtained by running:
names(temp)[temp == max(temp)]

75
Q

What command is used to get the variance of a variable in R?

A

The var() command is used to get the variance of a variable in R.

76
Q

What command is used to get the standard deviation of a variable in R?

A

The sd() command is used to get the standard deviation of a variable in R.

77
Q

What command is used to get a variable’s central tendency and range in R?

A

The summary() command can be used in R to get a variable’s central tendency and range.

78
Q

What command is used to perform a t-test in R?

A

The t.test(var1, var2) command is used to perform a t-test in R.

79
Q

What command is used to perform non-parametric tests in R?

A

There are four main non-parametric tests in R:

  • willcox.test(y~a) / willcox.test(y,x) / willcox.test(y1, y2, paired=TRUE)
  • kruskal.text(y~A)
  • friendman.test(y~A|B)
80
Q

What command is used in R to produce correlations?

A

The cor() command is used in R to produce correlations. It takes three variables, (x, use=, method=). X is the matrix or data frame, use specifies how to handle missing data, and method specifies the type of correlation (pearson, spearman, or kendall).

81
Q

What command is used in R to produce covariance?

A

The cov() function is used in R to produce covariance.

82
Q

What command is used in R to produce ANOVAs?

A

The aov() function is used in R to produce ANOVAs.

IT 497 - Research Methodology (3 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2024 Bold Learning Solutions. Terms and Conditions