Brainscape's Knowledge GenomeTM

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


See full index

Comps > Stats > Flashcards

Stats Flashcards

1
Q

The more heterogenous the distribution of scores,

  • The smaller the standard deviation
  • The larger the standard deviation
  • The more the distribution tends to be skewed
  • The greater the chance of excluding extreme scores
  • The larger the mean
A

The larger the standard deviation

The more heterogenous (different, variance, all over the place) the distribution of scores, the larger the standard deviation. Also, SD= square root of the variance, and similarly Variance = SD squared.

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

For any normal distribution, the 50th percentile corresponds to a z score of

  • 0
  • 50
  • 100
  • +/-1
  • None of these
A

0

Know the normal distribution curve, the very middle is 50%.

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

Other things being equal, an alpha level of .05 should lead to a rejection of the null hypothesis

  • More often than when alpha is set at .01
  • More often than when alpha is set at .10
  • Less often than when alpha is set at .01
  • You cannot tell from the information provided
A

More often than when alpha is set at .01

Alpha is set at .05, is less strict allows for more than acceptance of findings, .01 is more strict, effect has to be larger, stronger to be accepted.

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

*** Rejecting Ho when in fact it should have been accepted, causes

  • Standard Error
  • Sampling Error
  • Omega error
  • Type I error
  • Type II error
A

Type I error

Type I error: thinking you found something (rejecting the null) when in fact you havn’t found something.

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

*** When a difference is considered “statistically significant,” the most accurate interpretation of the finding is that the difference is probably

  • Meaningful and Profound
  • Due to chance and chance alone
  • Not due to chance alone
  • Not evident in the population
  • None of these
A

Not due to chance alone

Statistically significant: the findings are not likely due to chance alone (because there is always chance in everything)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

*** The power of a statistical test, such as the independent t-test, can be increased by

  • Increasing the sample size
  • Increasing the accepted alpha level (making alpha numerically larger)
  • Using a one-tailed test
  • All of the above
A

Increasing the sample size
Increasing the accepted alpha level (making alpha numerically larger)
Using a one-tailed test
ALL OF THE ABOVE

To increase power you have to do the harder stuff (get a larger sample, use one tailed instead of two tailed)

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

The only time the standard deviation may equal zero, is when

  • The range is less than 50
  • Every score in the distribution is the same
  • The range is negative
  • None of these, since the standard deviation may never be equal to zero
A

Every score in the distribution is the same

For the mean and SD to equal 1, all of the scores in the distribution have to be the same.

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

A scatter plat on which the array of points goes from upper left to lower right indicates

  • A positive correlation
  • A negative correlation
  • A zero correlation
  • Any of the above (a, b, or c) depending on the strength of the correlation
  • None of the above (a, b, or c)
A

A negative correlation

Visualize, over time, things are decreasing, so negative

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

*** When beta is increased (becomes numerically larger)

  • Power has decreased
  • Power has increased
  • The likelihood of rejecting the null hypothesis is maximized
  • The likelihood of obtaining a significant difference is maximized
  • The possibility of Type I error also is increased
A

Power has decreased

Power =1-B

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

The T score distribution always assumes a

  • Mean of 50, and a SD of 10
  • Mean of 100 and a SD of 15
  • Mean of 100 and a SD of 10
  • Mean of 0 and a SD of 1
  • None of the above
A

Mean of 50, and a SD of 10

T score= a mean of 50, and a SD of 10 (think of assessment measures)

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

The mean of a given normal distribution of raw score is 200 with a standard deviation of 25. What percentage of scores is above 225?

  • 5% of the scores
  • 10% of the scores
  • 16% of the scores
  • 25% of the scores
  • 34% of the scores
  • There is not enough information to tell
A

16% of the scores

Know the normal distribution curve; percentages under SD 1 to SD 3= 16%

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

If the standard deviation for a particular distribution is 6, then the variance equals

  • 2
  • 3
  • 6
  • 12
  • 36
A

36

Variance = SD squared

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

*** The alpha level states the probability of being wrong when

  • The null hypothesis is accepted
  • The null hypothesis is rejected
  • The alternative hypothesis is rejected
  • The sample means are assumed to be equal
  • The effect size is too small to detect
A

The null hypothesis is rejected

Type I error (rejecting the null hypothesis) goes with alpha

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

Of the following correlation coefficients, which expresses the strongest association?

  • .76
  • .45
  • .00
  • -.37
  • -.95
A

-.95

Can be positive or negative, the higher the number the better

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

*** Which of the following terms best indicates a process variable (X) through which one variable (A) influences another (B) (e.g., it explains how A influences B)?

  • Risk Variable
  • Associated Variable
  • Mediator Variable
  • Moderator Variable
  • Correlate
A

Mediator Variable

Mediator variable is how/why, moderator variable is strength?

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

Which of the following types of data uses numbers or letters to label categories?

  • Nominal
  • Ordinal
  • Interval
  • Rank
  • Ratio
A

Nominal

Pinot NOIR (the acronym progressively becomes more about numbers, quantitative)

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

A correlation coefficient of .80 yields a coefficient of determination of

  • 80%
  • 40%
  • 64%
  • 8%
  • None of the above
A

64%

Correlation Coefficient squared= coefficient of determination

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

As the sample size increases, degrees of freedom

  • Remain the same
  • Increase
  • Decrease
  • Are completely unaffected
A

Increase

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

*** When the alternative hypothesis is accepted

  • The “chance” explanation is completely ruled out
  • The “chance” explanation, though not ruled out, is judged improbable
  • The “chance” explanation is fully accepted
  • The “chance” explanation becomes irrelevant
  • Beta should be carefully considered as an explanation
A

The “chance” explanation, though not ruled out, is judged improbable

Chance is never ruled out, but seen to be improbable

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

Which of the following does NOT contribute to a small effect size?

  • The heterogeneous sample
  • A large standard deviation among the scores on the dependent variable
  • Variable experimental procedures
  • Use of unreliable measures
  • Use of uncontrolled multiple comparisons
A

Use of uncontrolled multiple comparisons

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

When the standard deviation of the entire distribution of random sample means has been calculated, the resulting value is called

  • The deviation score
  • The parameter mean
  • The standard error of the mean
  • Sampling error
  • The mean square value
A

The standard error of the mean

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

*** The smaller the effect size, the greater the likelihood

a. Of accepting the null hypothesis
b. Of rejecting the null hypothesis
c. Of committing a Type I error
d. Of committing Type II error
Both a and d
Both b and c

A

Both a and d

a. Of accepting the null hypothesis
d. Of committing Type II error

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

The mean of a given normal distribution of raw scores is 200 with a standard deviation of 25. Between the raw scores of 150 and 250, there must be

  • 50% of scores
  • 68% of scores
  • 80% of scores
  • 84% of scores
  • 95% of the scores
  • There is not enough information to tell
A

95% of the scores

Know the normal distribution curve; between SD of -2 and SD of 2, lies 95% of the scores

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

A researcher wishes to test the hypothesis that there is a positive association between IQ and musical ability. A random sample of 10 subjects was selected. The subjects were rank-ordered in terms of musical ability. A random sample of 10 subjects was selected. The subjects were rank-ordered in terms of musical ability. Then each subject was given an IQ test. Which of the following statistical tests is most appropriate to determine whether there is a significant relationship between IQ and musical ability in this situation?

  • Spearman r
  • Pearson r
  • Fisher’s z
  • Point-biserial correlation
  • Phi coefficient
A

Spearman r

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

*** Which of the following statements is NOT true?

  • The use of multiple comparisons contributes to an experiment-wise error rate that may be too high
  • The experiment-wise error rate is typically higher than the alpha level set for an individual comparison
  • The experiment-wise error rate is typically lower than the alpha level set for an individual comparison.
  • An elevated experiment-wise error rate contributes to Type I error
  • Problems with the experiment-wise error rate may cause the researcher to erroneously reject the null hypothesis
A

The experiment-wise error rate is typically lower than the alpha level set for an individual comparison.

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

When a distribution shows a large majority of high scores and a few very low scores, the distribution is said to be

  • Skewed to the right
  • Skewed to the left
  • Skewed to the middle
  • Bimodal
  • None of the above
A

Skewed to the left

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

A Pearson r value of -.16

  • Can never be significant
  • Must always be significant
  • Suggests that a slight positive relationship exists between two variables
  • Cannot be evaluated for significance unless df is known
A

Cannot be evaluated for significance unless df is known

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

Once an F ratio is found to be significant, then

  • Each sample mean must be significantly different form each of the other sample means
  • At least one of the sample means must be significantly different from all the other sample means
  • At least one of the sample means must be significantly different from at least one of the other sample means.
  • Each sample mean deviates from the overall mean by the same amount
  • None of the above
A

At least one of the sample means must be significantly different from at least one of the other sample means

F ratio= one sample mean is different from other sample means

29
Q

A researcher obtains a 2 X 2 chi square value of 31.30. The appropriate degrees of freedom indicated critical table values of 31.41 at .05 and 37.57 at .01. The researcher should

  • Accept Ho
  • Reject Ho at .05
  • Reject Ho at .01
  • Only test the a priori hypothesis
A

Accept Ho

Did not exceed the critical values, we therefore accept the null stating that nothing happened/our hypothesis (the alternative) wasn’t proven

30
Q

In order to calculate an ANOVA, there must be

  • More than one sample of scores
  • Interval or ratio data
  • Generally homogenous variances among the sample groups
  • A normal distribution in the population from which the samples were selected
  • All of the above
A
  • More than one sample of scores
  • Interval or ratio data
  • Generally homogenous variances among the sample groups
  • A normal distribution in the population from which the samples were selected
  • ALL OF THE ABOVE

ANOVA= analysis of variance (between more than one sample), you would therefore have to have more than one sample of scores. Requires I/R data (think numbers). The individuals scores need to be more similar to others in their group, similar to other groups, so that we can accurately/better compare them.

31
Q

Which of the following types of multiple regression allow(s) the researcher to determine in advance, the order in which specific predictor variables are entered into the regression equation?

  • Simple simultaneous multiple regression
  • Hierarchical multiple regression
  • Forward step-wise regression
  • Backward stepwise regression
  • All of the above
A

Hierarchical multiple regression

32
Q

When calculating an ANOVA, the individual differences among subjects not attributable to the independent variable are most clearly shown in the value of the

  • MS total
  • MS within
  • MS between
  • The F ratio
  • Degrees of freedom (df)
A

MS within

33
Q

A researcher wants to compare the average level of externalizing symptoms on the CBCL separately for boys and girls who are in one of the 3 following programs; a special education program, a regular education curriculum, and a self-contained program for gifted and talented. (Note that this question has two parts, A and B, as indicated below – the B part will be extra credit).

A. Which label best describes the type of analyses that would be required?

  • One way ANOVA
  • 2 X 3 ANOVA
  • 1 X 3 ANOVA
  • 1 X 3 Chi square
  • 2 X 3 Chi square

B. How many cells would be required to depict the design described above, in question 6?

  • 2
  • 3
  • 4
  • 5
  • 6
A

A. Answer: 2 X 3 ANOVA

B. Answer: 6

a. boys and girls (2), number of programs (3), therefore a 2 x 3
b. 2 multiplied by 3 = 6

34
Q

Which of the following tests is NOT used as a post hoc analysis following an ANOVA?

  • Tukey’s HSD
  • Newman-Keul’s Test
  • Multiple t-tests with a Bonferroni’s adjustment
  • Cohen’s kappa
A

Cohen’s kappa

35
Q

Subjects in a randomized controlled trial comparing two treatments are assessed with a variety of measures include a single Likert scale post-test item that asks them to rate how much they liked the treatment. The appropriate statistical test for the effect of treatment on this item is

  • A factorial ANOVA
  • Wilcoxon T-test
  • A multivariate analysis of variance (MANOVA)
  • Mann-Whitney U
  • Kruskal-Wallis H
A

Mann-Whitney U

36
Q

The concept of “within variability” is an indication of

  • How far the sample means deviate from each other
  • How far the scores deviate from their own sample means
  • How far the scores in all groups deviate from the total mean
  • How far each sample mean deviates from the total mean
A

How far the scores deviate from their own sample means

Within variability: within a group, how much the scores vary of the members of that group

37
Q

Which of the following was NOT discussed as a measure of interrater reliability?

  • Cohen’s kappa
  • Pearson product moment correlation
  • Percent of agreement between 2 observers
  • Paired t-test
  • None of the above
A

Paired t-test

38
Q

Tukey’s HSD is used in order to

  • Establish the significance of the overall F ratio
  • Determine inter-rater reliability
  • Reduce the beta error
  • Determine inter-item reliability
  • None of the above
A

None of the above

39
Q

The less overlap there is among the various sample-group distributions,

  • The larger the F ratio
  • The smaller the F ratio
  • The greater the degrees of freedom
  • The greater the r value
  • The smaller the t value
  • None of the above
A

The larger the F ratio

Less overlap= more variance= larger F ratio

40
Q

An F ratio of 5, tells us specifically that the variance (MS) between groups is

  • 5 times smaller than the variance within
  • 5 times larger than the variance within
  • 5 times larger than the total variance
  • 10 times larger than the variance within
  • 10 times smaller than the variance within
A

5 times larger than the variance within

F= MSb/MSw

41
Q

The chi square is a

  • Parametric test of mean differences
  • Nonparametric test of frequency differences
  • Parametric test of frequency differences
  • Nonparametric test of mean differences
  • Nonparametric test of rank differences
A

Nonparametric test of frequency differences

42
Q

If I want to compare the percentage of “fast-track” (e.g. 4-year-track) students at ASPP who are accepted for internship with the percentage of “5-year-track” students who are accepted for internship, the most appropriate type of statistical analyses to do is

  • 1-way ANOVA
  • 2 X 2 ANOVA
  • 2 X 2 (Xsquared)
  • t test for independent samples
  • Partial correlation
A

2 X 2 (X squared)

I cant figure out how to make and X with a little 2 next to it…

43
Q

The F ratio is generated by

  • Dividing the MS between by the MS within
  • Dividing the MS within by the MS between
  • Dividing the SS between by the df between
  • Dividing the SS between by the SS within
  • None of the above
A

Dividing the MS between by the MS within

F= MSb/MSw

44
Q

The null hypothesis for chi square states that the

  • Means for each group are equal
  • Fo’s are equal to the Fe’s
  • Population mean is equal to the sample mean
  • Means for each group are equal to the Fo’s
  • None of the above
A

Fo’s are equal to the Fe’s

Null hypothesis for Chi square: Fo=Fe

45
Q

Which of the following is NOT a requirement for a two-sample t-test?

  • Sample scores must be provided in interval or ratio data
  • The two samples must be dependent on each other
  • Standard deviations of the two samples must be similar
  • Must be normal distributions
  • None of the above
A

The two samples must be dependent on each other

46
Q

Which of the following is used to prevent a 2 X 2 chi square value from becoming inflated, thus yielding excessive alpha error?

  • Fisher’s z test
  • Wilcoxon test
  • Bonferroni correction
  • Tukey’s HSD
  • None of the Above
A

None of the Above

47
Q

The one-way ANOVA may only be used when

  • The direction of the difference is specified
  • The F ratio is significant
  • There are more than 3 sample groups
  • There is only one independent variable
  • There are 2 or more dependent variables
A

There is only one independent variable

48
Q

When calculating a higher order (factorial) ANOVA, the main effects are found from the

  • Rows
  • Columns
  • Both of these
  • Neither of these
A

Both of these (rows and columns)

49
Q

A measure of the internal consistency of a measure is

  • Cohen’s Kappa
  • Cronbach’s alpha
  • Wilcoxon Test
  • Tukey’s HSD
  • Newman Keul’s Test
  • None of the Above
A

Cronbach’s alpha

50
Q

Reasons discussed in class for why the SSRI literature may be misleading about efficacy include

a. Journal editors and reviewers are biased
b. Only statistically significant findings are likely to be submitted or published
c. Researcher bias is reflected in publications
d. Published studies do not reflect the rate at which hypotheses are tested
e. All of the above
f. b and d

A

Both b and d

b. Only statistically significant findings are likely to be submitted or published
d. Published studies do not reflect the rate at which hypotheses are tested

51
Q

Strengths of a single group longitudinal design include

  • Determining the temporal relationship between variables
  • Cohort Effect
  • Attrition
  • All of the above
  • None of the above
A

Determining the temporal relationship between variables

52
Q

Reactivity is a potential threat primarily to which validity?

  • Statistical
  • Construct
  • External
  • Internal
  • None of the above
A

External

53
Q

A researcher is interested in the risk that parental separation/divorce poses for the development of dysthymia and/or major depression among children. She identifies two groups of first grade children. The first group (n = 50) have experienced paternal separation or divorce during the last six months, and the second group (n = 50) have not experienced paternal separation/divorce. The researcher follows them annually for 5 years, assessing levels of depressive symptoms among the children. At the end of the 5-year study, she compares the data to determine whether the two groups differ in their levels of depressive symptoms. Which type of research design is used in this study?

  • Single group cohort design
  • Multigroup cohort design
  • Accelerated, multicohort, longitudinal design
  • Cross-sectional case control design
  • Retrospective case control design
A

Multigroup cohort design

54
Q

A correlation coefficient of .70 yields a coefficient of determination of

  • 49%
  • 70%
  • 35%
  • 140%
  • None of the above
A

49%

correlation coefficient squared= coefficient of determiation

55
Q

A researcher wishes to test the hypothesis that there is a positive association between scholastic ability and artistic skill. A random sample of 10 subjects was selected. The subjects were rank-ordered in terms of artistic skill. Then each subject was given an achievement test. Which of the following statistical tests is most appropriate to determine whether there is a significant relationship between scholastic ability and artistic skill in this situation?

  • Spearman r
  • Pearson r
  • Fisher’s z
  • Point-biserial correlation
  • Phi coefficent
A

Spearman r

56
Q

A researcher is interested in exploring any associates that may exist between childhood stressors and the diagnosis of antisocial personality disorder in adulthood. He identified one group of adults who have been diagnosed with antisocial personality disorder and another group of adults who have been diagnosed with other personality disorders. He then administered a structured interview designed to elicit perceptions of a variety of childhood stressors in both groups. This research design is best described as

  • Cross-sectional case-control design
  • Retrospective case-control design
  • Quasi experimental post test design
  • Single group cohort design
  • Multigroup cohort design
  • Pre-experimental post test design
A

Retrospective case-control design

57
Q

A researcher is interested in whether a new training technique affects reading achievement scores. If the group chosen for the study is already high in reading achievement levels at pretest, the lack of a significant increase in scores at posttest is mostly likely due to the

  • Researcher having chosen the .05 rather than the .01 level of statistical significance
  • Halo effect
  • Hawthorne effect
  • Experimenter expectancy
  • Ceiling effect
A

Ceiling effect

58
Q

For the quasi-experimental design, the following are all true, except:

  • The independent variable is manipulated, though the experimenter may not be able to control its administration
  • The independent variable identifies different treatment conditions
  • The participants are randomly assigned to the groups
  • There is a comparison group
A

The participants are randomly assigned to the groups

59
Q

When investigating a new treatment for an anxiety symptom for which the distributions for the untreated outpatient and normal populations do not overlap, a reasonable measure of clinical significant improvement for each patient would be a level of functioning that is

a. Improved 2 SD beyond mean of clinical population
b. Improved to be within 2 SD of mean of normal population
c. Is improved so that mean is closer to mean of normal than clinical population
d. Either b. or secondarily c., depending upon the difficulty of achieving b.
e. None of the above - clinical significance cannot be applied to measures of a symptom

A

d. Either b. or secondarily c., depending upon the difficulty of achieving b.

b. Improved to be within 2 SD of mean of normal population
c. Is improved so that mean is closer to mean of normal than clinical population

60
Q

After demonstrating that an overall treatment package, multisystemic therapy yields significant positive benefits in reducing acting out behaviors of adolescents, the researcher sets out to determine which components of the treatment are critical to the treatment approach. She compares the scores of one sample who receive the typical multisystemic therapy program with those of another sample who receive multisystemic therapy without the patent intervention that is typically included as a part of the treatment package. Which of the following terms best describes this type of study?

  • Dismantling
  • Constructive
  • Parametric
  • Comparative treatment
  • Client-therapist variation
A

Dismantling

61
Q

When conducting psychological research studies, some typical problems have NOT included

  • Mainstream psychologists not measuring cultural variables
  • Cross-cultural psychologists comparing behavioral phenomena without measuring cultural variables
  • Assumptions that mainstream findings were applicable to individuals everywhere
  • Cross-cultural psychologists ignoring the heterogeneity of nations or ethnic groups
  • None of the above
A

None of the above

62
Q

The disadvantages of experimental multi-treatment (not factorial) designs do NOT include

  • Ceiling effects
  • Order effects
  • Floor effects
  • Sequence effects
  • None of the above
A

None of the above

63
Q

Typical control group types do NOT include

  • Yoked
  • No treatment
  • Attention-placebo
  • Constructive
  • Waiting list
A

Constructive

64
Q

Based on lecture, what has NOT been mentioned as a biasing factor in the drug literature?

  • Consulting contracts
  • Bribes
  • Authorship of drug company produced papers
  • Free drug samples
  • Free trips, dinners, and continuing education
  • None of the above
A

Bribes

65
Q

In a multiple baseline design, the baselines can include all BUT

  • Different behaviors
  • Same behavior in different settings
  • Same behavior when intervention is repeatedly turned on and off
  • Same behavior in different individuals
  • None of the above
A

Same behavior when intervention is repeatedly turned on and off

66
Q

Single-subject experimental designs include all of the following except

  • Changing criterion designs
  • ABAB designs
  • Multiple-baseline designs
  • Visual inspection designs
  • None of the above
A

Visual inspection designs

67
Q

Common problems with research questions from the perspective of gay men, lesbians, bisexual, and transgender (LGBT) people have included all except:

A. Ignoring or denying the existence of LGBT people
B. Implicitly assuming that observed characteristics are caused by subjects sexual orientation
C. Devaluing or stigmatizing LGBT people
D. Reflecting cultural stereotypes of LGBT people
E. None of the above

A

None of the above

68
Q

A researcher wishes to test the hypothesis that there is an association between scholastic ability and punctuality. For a class of 20 students, the students were rank-ordered in terms of arrival at the class a week before a quiz. As one might expect, there were early birds and some students somewhat late, with a block of students in between coming in at the same time, resulting in rank ties for this middle group. Then each students percentage grade on the quiz the following week was recorded. Which of the following statistical tests is most appropriate to determine whether there is a significant relationship between scholastic ability and punctuality in this situation?

A. Spearman r
B. Pearson r
C. Kendall's tau
D. Point-bisexual correlation
E. Phi coefficient
A

Kendall’s tau

Comps (3 decks)

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