Research and Analysis Exam

Question 1

What are the ethical standards that research must adhere to? The research must respect..

Answer 1

Dignity
Rights
Safety
Wellbeing
of the people who take part

Question 2

What does research communication mean in ethics?

Answer 2

Research communication must be plagiarism free and you must accurately represent your results

Question 3

What does informed consent entail?

Answer 3

1. What the study will be about
2. The risks / benefits of taking part
3. Your supervisor’s contact info and the institution’s ethics approval number
4. Making sure that the data is kept confidential and that they are free to stop filling in the survey at any point for any reason
5. Should give them at least 24 hrs to decide and opportunity to ask questions

Question 4

What is data pseu-donymisation?

Answer 4

Data pseu-donymisation is an alterative method where you replace identifiable info of participants with fake (pseudonymous) identifiers.

Data can still be linked to participants but its harder to do so because you separate personal info from the study data

Question 5

Giving each participant a random 3 digit number is an example of what ethical issue?

Answer 5

Anonymity.

You separate their personally identifying info from their research data and include the participant numbers in both files. Data can only be accessed by participants and researchers

Question 6

What are the 4 factors of potential for harm?

Answer 6

Psychological

Social harm

Physical harm

Legal harm

Question 7

When can secondary data analysis be used?

Answer 7

If the data is purely quantitative

Question 8

Theory vs. Data

What is hypothesis induction? What is the opposite of this?

Answer 8

Hypothesis induction is when a theory or hypothesis is formed using data collect from research

Deduction is when a theory is being tested and used to make predictions and hypotheses about the data

Question 9

What is truncation in regard to literature searching?

Answer 9

Truncation, stemming, is used to broaden your search to include various word endings and spellings by using an “*” or other symbols at the end of the root word

Example:
child* = child, childs, children, childrens, childhood

Question 10

What is “wildcarding” in literature searching?

Answer 10

Wildcard, allows you to search for variations within the word (useful for picking up terms for UK/US variations) e.g. wom?n

Question 11

What is adjacency / proximity in literature searching?

Answer 11

Allows you to decide how far apart you want two key words to be

For example: The search ‘mobile app$ adj3 effective$’ will return results in which “mobile app$” and “effective” are within three words of each other. To use the adjacent function use the feature adj within your search followed by the number of words you wish to limit it by.

Question 12

What does it mean to have different levels of Independent Variables?

Answer 12

It means that there can be different categories of IV’s, called levels, conditions or treatments

Question 13

Can there be multiple Independent variables?

Answer 13

Yes. You can have multiple IV’s but only belong to one level for each

Ex. Vaccine study participants have 2 levels (vaccinated or not)

Ex. Your height, weight and BP - you can have multiple of these

Question 14

What is an example of categorical, ordered data?

Answer 14

Year of Undergraduate study (Yr 1-3)

Question 15

What is an example of categorical, not ordered data?

Answer 15

Which degree program you’re in

Question 16

What is the difference between nominal and ordinal data?

Answer 16

Nominal data is categorical and cannot be counted

E.g. Gender, occupation etc.

Ordinal data however, is categorical and can be counted, but not added or subtracted

E.g. satisfaction rating (extremely dissatisfied to extremely satisfied), education level (high school to PhD), spicy (mild to hell)

Question 17

What is interval data?

Answer 17

Interval data can be ordered, but you cannot compute a ratio between two values because no meaningful zero exists

Examples: Exam mark, date, year

Question 18

What is a ratio data type?

Answer 18

Ratio data is interval data that can have a ratio between the two because it has a meaningful zero

Example:

Distance, height, annual income, number of successes etc.

Question 19

Number of years in education is an example of which type of data?

Answer 19

Ratio: Even if the same unit “year” is being used, now you can imagine
calculating the ratio between two values, e.g. “my years of education are twice
yours”, because there exist the meaningful zero.

Question 20

Blood type is an example of which type of data?

Answer 20

Nominal: Blood types are categories but cannot be ordered.

Question 21

Year (in Gregorian calendar, e.g. 2002) is an example of which type of
data?

Answer 21

Interval: Years can be ordered and have measurable & fixed intervals. Also
their intervals, i.e. difference between years, can be calculated. However, their
ratio cannot be taken because there is no meaningful zero in the years. In
other words, it makes no sense to double an year, e.g. twice of 1988

Question 22

A response on a questionnaire about service quality between 1
(extremely unhappy) and 7 (extremely happy) is an example of which
type of data?

Answer 22

Ordinal: Response can be ordered but their intervals cannot be
measured/quantified

Question 23

What is the median

Answer 23

Middle value. Cannot be obtained in nominal data

Question 24

How do you calculate the mean of a set of data?

Answer 24

Sum of all the values divided by the number of values

The mean can be greatly affected by an outlier

Question 25

Sections of data containing the same amount of data is called what?

Answer 25

Quantiles.
If there are 4, they are called quartiles
If there are 100, they are called percentiles

Question 26

What is standard deviation?

Answer 26

The standard distance from the mean which is the square root of variance

Question 27

What is a Z-score?

Answer 27

A Z-score is ratio with respect to Standard Deviation that enables fair comparison of deviations

It is the value minus the mean divided by the SD

Ex. Mary’s height is 5.6 ft and is 0.3 taller than the mean. Therefore, (5.6-0.3) / 0.3 = 1.0

A Z-score of 0 means the data point is exactly at the mean of the dataset.

A positive Z-score indicates that the data point is above the mean of the dataset.The larger the positive Z-score, the farther above the mean the data point is.
For example, a Z-score of 2 means the data point is 2 standard deviations above the mean.

The opposite is true for negative Z-scores, it would mean that the value is below the mean

In Jamovi, units are the SD plus or minus from the mean

Question 28

What does positive and negative skewness of data mean?

Answer 28

Positive skewness means that the data is shifted towards the left

Negative skewness means that the data is shifted towards the right

If skewness is zero, the data is not shifted and is in the centre/symmetrical

Question 29

What does the kurtosis of data mean?

Answer 29

Kurtosis measures the “sharpness” of data or how thin it is. it is always positive

Question 30

At what value can you exclude an outlier from a data set?

Answer 30

If the Z-score is greater than 3, or less than -3

i.e. when the distance from the mean is more than 3 SD’s or less than -3 SD’s

OR

When using Inter-Quartile Range (IQR): the width between the 1st and 3rd quartile, if the outlier is greater than 1.5 IQR above the 3rd quartile or less than 1.5 IQR below the 1st quartile, it can be excluded from the dataset

Question 32

In the binomial function:
Bi ( A / B, q ) = 0.25 what do each of the numbers mean?

Answer 32

That would mean that the probability of obtaining A on the condition of B. “q” would mean the probability of success rate.

For example: Bi ( 3 / 3, 0.5) = 0.125
means that the probability of obtaining 3 heads in three trials where the expected probability of getting heads is 50% is 0.125

Question 33

What is a binomial distribution used for?

Answer 33

When there is a probability distribution of two choices. Using a BD you can assess how likely that an event (i.e. k heads in n tosses) happens by chance.

Question 34

When one calculates the probability that the value falls in a certain range, what is this called?

Answer 34

Cumulative probability

Example: What is the probability of getting less than 40 heads on 100 times