Deck 12: Analysing Quantitative Data

Question 1

Processing and analysing information turns it into useful data.

Answer 1

False

Question 2

What do graphs, charts and statistics help researchers do with their data?

A) Present their findings
B) Describe relationships between variables
C) Examine trends in the data
D) All of these

Answer 2

All of these

Question 3

Which of these is NOT a type of quantitative data?

A) Narrative accounts
B) House prices
C) Frequency of employee sick days
D) Percentage of satisfied customers

Answer 3

Narrative accounts

Question 4

Which of the following is the most important element of quantitative data analysis in a student project?

A) Complex modelling
B) Using appropriate statistics
C) Understanding formulae
D) Using a large number of statistical tests

Question 5

Data whose values cannot be measured numerically but can be classified into descriptive sets or placed in rank order are called ________ data.

A) Categorical
B) Qualitative
C) Interval
D) Ratio

Question 6

Which of the following is NOT a type of categorical data?

A) Dichotomous
B) Nominal
C) Ordinal
D) Discrete

Question 7

Numerical data for which you cannot state the relative difference between values are called ________ data.

A) Ratio
B) Ordinal
C) Interval
D) Nominal

Question 8

"A travel company decides to halve its advertising spend from ò.5 million to ñ.25 million a year during a recession." What type of data are being described here?

A) Ratio
B) Interval
C) Ordinal
D) Nominal

Question 9

What is a 'data matrix'?

A) A coding schedule
B) A plan of analysis
C) A table of cases and variables
D) A list of data types

Question 10

Data is usually entered into quantitative analysis software by representing the response for each variable for each case by a series of codes.

Question 11

What is a 'code' in the context of quantitative data analysis of questionnaires?

A) An identifier for each member of the sample
B) A number used to represent a respondent's answer
C) A self-memo the researcher has made
D) None of the above

Question 12

Why is it a good idea to use an existing coding scheme for your data if one is available?

A) It allows you to compare your data to other surveys
B) It allows you to be more flexible
C) It means your responses are more accurate
D) It will ensure you use the correct statistical test

Question 13

Which of these is NOT a possible reason for missing data?

A) Data were not required from the respondent
B) Refusal to answer
C) Non-completion of an item implies an answer
D) An unclear answer

Question 14

Sam has included an open question on her questionnaire asking respondents to explain their reasons for choosing a brand of lager. This data are likely to be coded ________.

A) Before collection
B) During collection
C) Twice
D) After collection

Question 15

What is a 'codebook'?

A) An explanatory record of the labels you have assigned to responses
B) A pre-existing scheme from previously published research
C) The name of a software package
D) The marking criteria from your tutor

Question 16

Why might a researcher want to weight the cases in their dataset?

A) To make up for missing data
B) To avoid under-representation in stratified samples
C) To attach significance to the most important data
D) To increase response rates

Question 17

A good way to begin quantitative data analysis is to produce diagrams and tables for individual cases.

Question 18

Fiona has obtained data on the number of 2nd year students who have debts of less than õ000 and the number of students who have debts of õ000 or more. Which diagram or table would be best to show these data so that specific numbers can be read easily?

A) Histogram
B) Bar chart
C) Multiple line graph
D) Cross tabulation

Question 19

Alexis needs to identify the point at which falling sales revenues are matched by rising costs. Which diagram would be most appropriate for her to use?

A) Multiple line graph
B) Pie chart
C) Histogram
D) Any of these

Question 20

Which of the following diagrams is most suitable for showing precise highest and lowest values of continuous data?

A) Bar chart
B) Pictogram
C) Histogram
D) Line graph

Question 21

What is a 'scatter plot' most appropriate for showing?

A) Distribution of values for two variables
B) Relationships between cases for two variables
C) The interdependence of two variables
D) The proportional occurrence of two variables

Question 22

Proportions are more appropriately represented on a pictogram than a pie chart.

Question 23

Data which are 'positively skewed' have values concentrated on the ________ .

A) Right of the central point
B) Left of the central point
C) Top half of the curve
D) Bottom half of the curve

Question 24

"The pointedness or flatness of a distribution of data values compared with the normal distribution." What is being described here?

A) Kurtosis
B) Skewness
C) Frequency
D) Mode

Question 25

The boxed area of a 'box plot' depicts the _________.

A) Outliers
B) Median value
C) Highest and lowest values
D) Middle 50% of the data

Question 26

A multiple bar chart compares highest and lowest values of grouped variables, whereas a percentage component bar chart compares ________.

A) Average variables
B) Proportions of grouped variables
C) Trends and conjunctions
D) All of these

Question 27

The dependent variable should be plotted on the vertical axis of a graph.

Question 28

A negative relationship between two variables on a scatter graph is indicated by a slope of the imaginary line that ________.

A) Points down to the right
B) Points down to the left
C) Is horizontal
D) Is almost vertical

Question 29

The mode, median and mean are all measures of ________.

A) Kurtosis
B) Distribution
C) Central tendency
D) Coding

Question 30

Accepting the null-hypothesis means that there ________.

A) Is a significant relationship or difference between variables
B) Is no significant relationship or difference between variables
C) Is a significant difference between variables
D) Is no significant relationship between variables

Question 31

The Type I error concludes that you accept the null hypothesis although there is a significant relationship.

Question 32

To accept a null-hypothesis, the probability of your test statistic normally has to be higher than ________.

A) 0.01
B) 0.05
C) 0.1
D) 0.5

Question 33

The Type II error concludes that you reject the null hypothesis although there is no significant relationship.

Question 34

Descriptive statistics calculate the central tendency and dispersion of a variable.

Question 35

Which statistical test should you use to test the association between variables?

A) ANOVA
B) Kolmogorov-Smirnov test
C) Independent groups t-test
D) Chi square test

Question 36

Which statistical test should you use to assess the strength of a linear relationship between variables?

A) Correlation
B) ANOVA
C) Chi-square test
D) Regression

Question 37

Which statistical test should you use to assess the strength of a cause-and-effect relationship between variables or to predict a value?

A) Correlation
B) ANOVA
C) Chi-square test
D) Regression

Question 38

Rachel's research has found waiting times at a hospital clinic have a mean of 35 minutes. However, some times are busier than others and waiting times vary from 5 minutes one day to 3 hours on another. What statistic will be best to describe the extent to which the waiting times in her data differ from the mean?

A) Mean range
B) Standard deviation
C) Inter-quartile range
D) None of these

Question 39

Marianna wants to compare the relative extent to which sales of two companies products have changed over the past 10 years. She has obtained sales data from both companies for each of the past ten years. Which statistic will best allow her to do this?

A) Index numbers
B) Coefficient of variation
C) Index variation
D) Change index

Question 40

Seyma has created a table using categorical data from two questions in her questionnaire. The table has two rows and three columns. Which statistic should she use to test whether the responses to the two questions are associated?

A) Paired t test
B) Independent samples t test
C) Chi square
D) Phi

Question 41

A paired t test can be used where data for a quantifiable variable can be divided into two distinct groupings using a descriptive variable.

Question 42

ANOVA can be used when data for a quantifiable variable can be divided into three or more mutually exclusive groupings using a descriptive variable.

Question 43

Jemma is unsure whether to use a t test to see whether two groups are significantly related. Her ratio data, consisting of 120 cases, are skewed and the two groups are have different variances. She should ________.

A) Not use the t-test, it is only suitable for nominal data
B) Use the t test but make a note in her project that the assumptions of the test were not satisfied
C) Not use the t test, the assumptions of the test are not been satisfied
D) Use the t test, it is suitable for her data and the assumptions of the test have been satisfied

Question 44

Ben wishes to calculate the average gender of employees in his organisation. Which statistic should he use?

A) Mean
B) Median
C) Mode
D) Harmonic mean

Question 45

Doris has read a research paper in a refereed journal which states "within this organisation age is strongly correlated with gender (r = 0.7, p <0.01)". She should ________.

A) Accept the statement, correlation is the right statistic to use with these data types
B) Reject the statement, age and gender are never correlated
C) Reject the statement, correlation is the wrong statistic to use with these data types
D) Accept the statement, it is published in a referred journal

Question 46

What do you need to undertake to predict the value of a variable from two or more other variables?

A) Regression analysis
B) Correlation analysis
C) Prediction analysis
D) Variable analysis

Question 47

In the equation "y = a + bx", what are "y" and "a"?

A) y is the independent variable, a is the constant
B) y is the dependent variable, a is the constant
C) y is the independent variable, b is the coefficient
D) y is the dependent variable, b is the coefficient

Question 48

Dave wants to test whether those who responded to an employee questionnaire are representative of the total population of employees. He has data about occupation for both his respondents and the total population of employees which he can rank in order of seniority. Which statistical test would be most suitable?

A) Chi square test
B) Independent samples t test
C) Kolmogorov-Smirnov test
D) Cramer's V

Question 49

Which correlation coefficient is most appropriate if you have tied ranks in your data?

A) Pearson's r
B) Spearman's rho
C) Kendal's tau
D) Cronbach's alpha