PSYCHOLOGY EDEXCEL RESEARCH METHODS ||
WITH ERRORLESS SOLUTIONS 100%.
Importance of Level of Measurement correct answers decide on what descriptive statistics and
what inferential statistical test is appropriate for the data
nominal level of measurement correct answers - data organised categories with a measure of
frequency
-frequency data, counts
-mutually exclusive
-contains the least info
ordinal level of measurement correct answers data that are ordered in some way
-size order
-exact difference between each item is not known
-ranked or rating scales (but subjective)
-exact difference between each point is not known
interval level of measurement correct answers data made up of a scale of units that have equal
intervals
-shows how much difference between scores
-contains most info
significance level correct answers expresses the level of chance they are prepared to accept to be
sufficiently satisfied that the results obtained show a genuine difference in conditions or
correlation between variables
what's the traditional level of significance? correct answers 0.05
p<0.05 correct answers the probability of the results occurring due to chance is less than or equal
to 5% (0.05)
greater than chosen significance level? correct answers accept null hypothesis
When are inferential statistics used? correct answers to asses whether the difference or
correlation observed in our results are statistically significant
observed value correct answers value we calculate
,critical value correct answers value we read from the table
statements of significance include correct answers 1. the observed and critical values used and
which one was bigger
2. the sample size
3. the significance level used
4. whether the one or two tailed teat was used
5. which hypothesis was accepted
statement of significance example correct answers the observed value was 35 which was bigger
than the critical value which was 25. the sample size was 15 participants and the significance
level used was 5%.
a two tailed hypothesis was used and the null hypothesis was accepted at the end, stated "that
any difference between condition a and condition b was due to chance"
why might a researcher use p<0.05 ? correct answers -traditional and most frequently used
- offers reasonable balance between likelihood of making type 2 and type 2 errors
p<0.1 10% level correct answers the probability of the results having occurred by chance is less
than 10%
why might a researcher use p< 0.1 ? correct answers when there is no need for a strict level of
significance
beginning a new area of research
p<0.01 1% level correct answers the probability of the result having occurred by chance is less
than 1%
why might a researcher use p< 0.01 ? correct answers very strict level because any mistakes in
the research could have serious consequences
-may affect human health
type 1 error correct answers accepting the alternative hypotheses when the results are actually
due to chance (false positive)
type 2 error correct answers accepting the null hypothesis when the results were not due to
chance (false negative)
what error for 10%? correct answers easier to achieve significance
, more likely results are due to chance
type 1 error more likely
what error for 1%? correct answers difficult to achieve significance
type 2 error more likely
Non-parametric statistics correct answers statistical tests that can be carried out when dealing
with ordinal or nominal levels of measurement
Related Data correct answers Repeated measures and matched pairs
Unrelated data correct answers independent groups
Chi-square test correct answers unrelated= independent groups (diff)
nominal level of measurement
Mann-Whitney U test correct answers unrelated= independent groups (diff)
ordinal level of measurement
Wilcoxon Signed Ranks test correct answers related= repeated measures/matched pairs (diff)
ordinal level of measurement
Spearman's correlation coefficient test correct answers related from a correlational study
ordinal level of measurement
justifying choice of test correct answers 1. difference or correlation and why
2. level of measurement and why
3. Unrelated or related and why
justifying choice of test example correct answers Wilcoxon signed ranks:
a test of difference was required because
the data from the study would be at the ordinal level of measurement. this is because the memory
scores were out of 20 and could be ranked
the design produced related data because the pair of data is from one rat so its repeated measures
WITH ERRORLESS SOLUTIONS 100%.
Importance of Level of Measurement correct answers decide on what descriptive statistics and
what inferential statistical test is appropriate for the data
nominal level of measurement correct answers - data organised categories with a measure of
frequency
-frequency data, counts
-mutually exclusive
-contains the least info
ordinal level of measurement correct answers data that are ordered in some way
-size order
-exact difference between each item is not known
-ranked or rating scales (but subjective)
-exact difference between each point is not known
interval level of measurement correct answers data made up of a scale of units that have equal
intervals
-shows how much difference between scores
-contains most info
significance level correct answers expresses the level of chance they are prepared to accept to be
sufficiently satisfied that the results obtained show a genuine difference in conditions or
correlation between variables
what's the traditional level of significance? correct answers 0.05
p<0.05 correct answers the probability of the results occurring due to chance is less than or equal
to 5% (0.05)
greater than chosen significance level? correct answers accept null hypothesis
When are inferential statistics used? correct answers to asses whether the difference or
correlation observed in our results are statistically significant
observed value correct answers value we calculate
,critical value correct answers value we read from the table
statements of significance include correct answers 1. the observed and critical values used and
which one was bigger
2. the sample size
3. the significance level used
4. whether the one or two tailed teat was used
5. which hypothesis was accepted
statement of significance example correct answers the observed value was 35 which was bigger
than the critical value which was 25. the sample size was 15 participants and the significance
level used was 5%.
a two tailed hypothesis was used and the null hypothesis was accepted at the end, stated "that
any difference between condition a and condition b was due to chance"
why might a researcher use p<0.05 ? correct answers -traditional and most frequently used
- offers reasonable balance between likelihood of making type 2 and type 2 errors
p<0.1 10% level correct answers the probability of the results having occurred by chance is less
than 10%
why might a researcher use p< 0.1 ? correct answers when there is no need for a strict level of
significance
beginning a new area of research
p<0.01 1% level correct answers the probability of the result having occurred by chance is less
than 1%
why might a researcher use p< 0.01 ? correct answers very strict level because any mistakes in
the research could have serious consequences
-may affect human health
type 1 error correct answers accepting the alternative hypotheses when the results are actually
due to chance (false positive)
type 2 error correct answers accepting the null hypothesis when the results were not due to
chance (false negative)
what error for 10%? correct answers easier to achieve significance
, more likely results are due to chance
type 1 error more likely
what error for 1%? correct answers difficult to achieve significance
type 2 error more likely
Non-parametric statistics correct answers statistical tests that can be carried out when dealing
with ordinal or nominal levels of measurement
Related Data correct answers Repeated measures and matched pairs
Unrelated data correct answers independent groups
Chi-square test correct answers unrelated= independent groups (diff)
nominal level of measurement
Mann-Whitney U test correct answers unrelated= independent groups (diff)
ordinal level of measurement
Wilcoxon Signed Ranks test correct answers related= repeated measures/matched pairs (diff)
ordinal level of measurement
Spearman's correlation coefficient test correct answers related from a correlational study
ordinal level of measurement
justifying choice of test correct answers 1. difference or correlation and why
2. level of measurement and why
3. Unrelated or related and why
justifying choice of test example correct answers Wilcoxon signed ranks:
a test of difference was required because
the data from the study would be at the ordinal level of measurement. this is because the memory
scores were out of 20 and could be ranked
the design produced related data because the pair of data is from one rat so its repeated measures
