UNIT-4

Monitoring impact of Interventions: Tracking

impact interventions
Monitoring is a form of evaluation or assessment, though unlike outcome or impact
evaluation, it takes place shortly after an intervention has begun (formative
evaluation), throughout the course of an intervention (process evaluation) or midway
through the intervention (mid-term evaluation).

Monitoring is not an end in itself. Monitoring allows programmes to determine what is

and is not working well, so that adjustments can be made along the way. It allows
programmes to assess what is actually happening versus what was planned.

Monitoring allows programmes to:

 Implement remedial measures to get programmes back on track and

remain accountable to the expected results the programme is aiming
to achieve.
 Determine how funds should be distributed across the programme
activities.
 Collect information that can be used in the evaluation process.

When monitoring activities are not carried out directly by the decision-makers of the
programme it is crucial that the findings from those monitoring activities are
coordinated and fed back to them.

Information from monitoring activities can also be disseminated to different groups

outside of the organization which helps promote transparency and provides an
opportunity to obtain feedback from key stakeholders.

There are no standard monitoring tools and methods.  These will vary according to
the type of intervention and objectives outlined in the programme.  Examples of
monitoring methods include:

 Activity monitoring reports

 Record reviews from service provision (e.g. police reports, case
records, health intake forms and records, others)
 Exit interviews with clients (Survivors)
 Qualitative techniques to measure attitudes, knowledge, skills,
behavior and the experiences of survivors, service providers,
perpetrators and others that might be targeted in the intervention.
 Statistical reviews from administrative databases (i.e. in the health,
justice, interior sectors, shelters, social welfare offices and others)
  Other quantitative techniques.

Impact Evaluation

Impact evaluation assesses the changes that can be attributed to a particular

intervention, such as a project, program or policy, both the intended ones, as well as
ideally the unintended ones. In contrast to outcome monitoring, which examines
whether targets have been achieved, impact evaluation is structured to answer the
question: how would outcomes such as participants’ well-being have changed if the
intervention had not been undertaken? This involves counterfactual analysis, that is,
“A comparison between what actually happened and what would have happened in
the absence of the intervention.” Impact evaluations seek to answer cause-and-
effect questions. In other words, they look for the changes in outcome that are
directly attributable to a program.

Impact evaluation helps people answer key questions for evidence-based policy
making: what works, what doesn’t, where, why and for how much? It has received
increasing attention in policy making in recent years in the context of both developed
and developing countries. It is an important component of the armory of evaluation
tools and approaches and integral to global efforts to improve the effectiveness of
aid delivery and public spending more generally in improving living standards.
Originally more oriented towards evaluation of social sector programs in developing
countries, notably conditional cash transfers, impact evaluation is now being
increasingly applied in other areas such as the agriculture, energy and transport.

Impact evaluation measures the difference between what happened with the
programme and what would have happened without it.  It answers the question,
“How much (if any) of the change observed in the target population occurred
because of the programme or intervention?”

Rigorous research designs are needed for this level of evaluation. It is the most
complex and intensive type of evaluation, incorporating methods such as random
selection, control and comparison groups.

These methods serve to:

 Establish causal links or relationships between the activities carried

out and the desired outcomes.
 Identify and isolate any external factors that may influence the
desired outcomes.

For example, an impact evaluation of an initiative aimed at preventing sexual

assaults on women and girls in town x through infrastructural improvements (lighting,
more visible walkways, etc.) might also look at data from a comparison community
(town y) to assess whether reductions in the number of assaults seen at the end of
the programme could be attributed to those improvements. The aim is to isolate
other factors that might have influenced the reduction in assaults, such as training for
police or new legislation.
While impact evaluations may be considered the “gold standard” for monitoring and
evaluation, they are challenging and may not be feasible for many reasons,
including:

 They require a significant amount of resources and time, which many

organizations may not have.
 To be done properly, they also require the collection of data following
specific statistical methodology, over a period of time, from a range of
control and intervention groups, which may be difficult for some
groups.

Impact evaluations may not always be called for, or even appropriate for the needs
of most programmes and interventions looking to monitor and evaluate their
activities.

 To measure programme impact, an evaluation is typically conducted

at the start (known as a baseline) and again at the end (known as an
endline) of a programme. Measurements are also collected from a
control group with similar characteristics to the target population, but
that is not receiving the intervention so that the two can be
compared.
 Attributing changes in outcomes to a particular intervention requires
one to rule out all other possible explanations and control for all
external or confounding factors that may account for the results.

An evaluation of the impact of a campaign to raise awareness around the provisions

of a recently enacted law on violence against women for example would need to
incorporate:

 Baseline data on awareness of the law’s provisions prior to the

campaign for the intervention group;
 Endline data on awareness of the law’s provisions after the campaign
for the intervention group;
 Baseline data on awareness of the law’s provisions prior to the
campaign for a closely matched control group not exposed to the
campaign; and
 Endline data on awareness of the law’s provisions after the campaign
for a closely matched control group not exposed to the campaign.

Endline data allows the programme to see if there were external/ additional factors
that might influence the level of awareness among those not exposed to the
campaign. If the study design does not involve a randomly-assigned control group, it
is not possible to make a definitive statement regarding any differences in outcome
between areas with the programme and areas without the programme.
However, if statistically rigorous baseline studies with randomly assigned control
groups cannot be conducted, very useful and valid baseline information and endline
information can still be collected.

Evaluation requires technical expertise and training.  If the programme does not
maintain the capacity in-house, external evaluators should be hired to assist.

There are five key principles relating to internal validity (study design) and external
validity (generalizability) which rigorous impact evaluations should address:
confounding factors, selection bias, spillover effects, contamination, and impact
heterogeneity.

 Confounding occurs where certain factors, typically relating to

socioeconomic status, are correlated with exposure to the
intervention and, independent of exposure, are causally related to the
outcome of interest. Confounding factors are therefore alternate
explanations for an observed (possibly spurious) relationship
between intervention and outcome.
 Selection bias, a special case of confounding, occurs where
intervention participants are non-randomly drawn from the beneficiary
population, and the criteria determining selection are correlated with
outcomes. Unobserved factors, which are associated with access to
or participation in the intervention, and are causally related to the
outcome of interest, may lead to a spurious relationship between
intervention and outcome if unaccounted for. Self-selection occurs
where, for example, more able or organized individuals or
communities, who are more likely to have better outcomes of interest,
are also more likely to participate in the intervention. Endogenous
program selection occurs where individuals or communities are
chosen to participate because they are seen to be more likely to
benefit from the intervention. Ignoring confounding factors can lead to
a problem of omitted variable bias. In the special case of selection
bias, the endogeneity of the selection variables can cause
simultaneity bias.
 Spillover (referred to as contagion in the case of experimental
evaluations) occurs when members of the comparison (control) group
are affected by the intervention.
 Contamination occurs when members of treatment and/or
comparison groups have access to another intervention which also
affects the outcome of interest.
 Impact heterogeneity refers to differences in impact due by
beneficiary type and context. High quality impact evaluations will
assess the extent to which different groups (e.g., the disadvantaged)
benefit from an intervention as well as the potential effect of context
 on impact. The degree that results are generalizable will determine
the applicability of lessons learned for interventions in other contexts.

Meaning, Sources and Consequences of Stress

Stress is a general term applied to various psychologic (mental) and physiologic (bodily)
pressures experienced or felt by people throughout their lives.

Stress is defined as “a state of psychological and physiological imbalance resulting from the
disparity between situational demand and the individual’s ability and motivation to meet
those needs.”

Dr. Hans Selye, one of the leading authorities on the concept of stress, described stress as
“the rate of all wear and tear caused by life.”

Stress can be positive or negative:

 Stress is good when the situation offers an opportunity to a person to gain

something. It acts as a motivator for peak performance.
 Stress is negative when a person faces social, physical, organizational and
emotional problems.

Factors that are responsible for causing stress are called stressors.

Meaning, Sources and Consequences of Stress

Stress is a general term applied to various psychologic (mental) and physiologic (bodily)
pressures experienced or felt by people throughout their lives.

Stress is defined as “a state of psychological and physiological imbalance resulting from the
disparity between situational demand and the individual’s ability and motivation to meet
those needs.”

Dr. Hans Selye, one of the leading authorities on the concept of stress, described stress as
“the rate of all wear and tear caused by life.”

Stress can be positive or negative:

 Stress is good when the situation offers an opportunity to a person to gain

something. It acts as a motivator for peak performance.
 Stress is negative when a person faces social, physical, organizational and
emotional problems.

Factors that are responsible for causing stress are called stressors.
Causes of Stress

1. Career Concern

If an employee feels that he is very much behind in the corporate ladder, then he may
experience stress. If he seems that there are no opportunities for self-growth, he may
experience stress. Hence, unfulfilled career expectations are the significant source of stress.

2. Role Ambiguity

It occurs when the person doesn’t know what he is supposed to do, on the job. His tasks and
responsibilities are not clear. The employee is not sure what he is expected to do. It creates
confusion in the minds of the worker and results in stress.

3. Rotating Work Shifts

Stress may occur in those individuals who work on different work shifts. Employees may be
expected to work on day shift for some days and then on the night shift. That may create
problems in adjusting to the shift timings, and it can affect not only personal life but also
family life of the employee.

4. Role Conflict

It takes place when people have different expectations from the person performing a
particular role. It can also occur if the job is not as per expectation, or when a job demands a
certain type of behavior that is against the person’s moral values.

5. Occupational Demands

Some jobs are more demanding than others. Jobs that involve risk, and danger are more
stressful. Research findings indicate, job that cause stress needs constant monitoring of
equipments and devices, unpleasant physical conditions, making decisions, etc.

6. Lack of Participation in Decision-making

Many experienced employees feel that management should consult them on matters affecting
their jobs. In reality, the superiors hardly ask the concerned employees before taking a
decision. That develops a feeling of being neglected, which may lead to stress.

7. Work Overload

Excessive workload leads to stress as it puts a person under tremendous pressure. Work
overload may take two different forms:

 Qualitative work overload implies performing a job that is complicated or

beyond the employee’s capacity.
  Quantitative work overload is a result of many activities performed in a
prescribed time.

8. Work Underload

In this, case, too little work or very easy work is expected on the part of the employee. Doing
less work or jobs of routine and simple nature would lead to monotony and boredom, which
can lead to stress.

9. Poor Working Conditions

Employees may be subject to poor working conditions. It would include bad lighting and
ventilation, unhygienic sanitation facilities, excessive noise, and dust, presence of toxic
gasses and fumes, inadequate safety measures, etc. All these unpleasant conditions create
physiological and psychological imbalance in humans thereby causing stress.

10. Lack of Group Cohesiveness

Every group is characterized by its cohesiveness, although they differ widely in its degree.
Individuals experience stress when there is no unity among work group members. There are
mistrust, jealousy, frequent quarrels, etc., in groups and this lead to stress to employees.

11. Interpersonal and Intergroup Conflict

These conflicts take place due to differences in perceptions, attitudes, values and beliefs
between two or more individuals and groups. Such conflicts can be a source of stress for
group members.

12. Organizational Changes

When changes occur, people have to adapt to those changes, and this may cause stress. Stress
is higher when changes are significant or unusual like transfer or adoption of new technology.

13. Lack of Social Support

When individuals believe that they have the friendship and support of others at work, their
ability to cope with the effects of stress increases. If this kind of social support is not
available, then an employee experiences more stress.

The main causes of stress outside work or organization:

Causes of stress outside work

1. Civic Amenities
Poor civic amenities in the area in which one lives can be a cause of stress. Inadequate or lack
of public facilities like improper water supply, excessive noise or air pollution, lack of proper
transport facility can be quite stressful.

2. Life Changes

Life changes can bring stress to a person. Life changes can be slow or sudden. Gradual life
changes include getting older, and abrupt life changes include death or accident of a loved
one. Sudden life changes are highly stressful and very difficult to cope.

3. Frustration

Frustration is another cause of stress. It arises when goal-directed behavior gets blocked.
Management should attempt to remove barriers and help the employees to reach their goals.

4. Racial, Caste, and Religious Conflicts

Employees living in areas, which are often prone to conflicts among people based on
differences seen in their race, caste and religion do suffer more from stress. In the case of a
religion, the minorities and lower-caste people (especially in India) are subject to more stress.

5. Personality

We can classify people as ‘Type A’ and ‘Type B’.

The ‘Type A’ people:

 They feel guilty while relaxing.

 They get irritated by minor mistakes of self and others.
 They feel impatient and dislike waiting.
 They also multitask and prefer to do several things at one time.

The ‘Type B’ people are exactly opposite and hence are less affected by stress due to the
above factors.

6. Technological Changes

When there are any changes in technical fields, employees are under the constant fear of
losing jobs or need to adjust to new technologies. It can be a source of stress.

7. Career Changes

When a person suddenly switches to another job, he is under stress to shoulder new
responsibilities adequately. Under-promotion, over-promotion, demotion and transfers can
also cause stress.
Following habits can remarkably help to relieve stress:

 Regular meditation
 Physical exercise
 Balanced diet
 Focused thinking
 Control of anger
 Managing Depression
 Maintaining calmness in stressful situations
 Having a positive attitude towards life
 Harmony towards self and others, etc.

Evaluating Value change

Change in the workplace can make us uncomfortable as we step out of what we
normally do and attempt new ways of working. Adapting to change in the workplace
is often difficult because it’s too easy to stay accustomed to our habits, however
change is essential to many aspects of business and our working life.

Those companies that don’t change can stagnate and not develop. Our current world
is constantly evolving in terms of technology and many other challenging but
innovative ways. Having the ability to face these new challenges head-on will not
only develop the company as a whole but also those who work within it.

Value change refers to an adjustment made in the price of a stock, usually to reflect
the total number of outstanding shares issued and withheld currently by investors.

The daily changes made in shares held by various investors are influenced by the
daily changes seen in the demand and supply. These price changes are periodically
adjusted to go at par with the Market changes.

The Value change enables a group of stocks to be weighed equally so that they can
be evaluated easily. Moreover, the number of shares held by a certain investor can
change daily. And this is the reason why the total number of shares held by a
particular investor on a specific day needs to be updated every day to determine the
price changes.

Evaluating Value change

Change in the workplace can make us uncomfortable as we step out of what we
normally do and attempt new ways of working. Adapting to change in the workplace
is often difficult because it’s too easy to stay accustomed to our habits, however
change is essential to many aspects of business and our working life.
Those companies that don’t change can stagnate and not develop. Our current world
is constantly evolving in terms of technology and many other challenging but
innovative ways. Having the ability to face these new challenges head-on will not
only develop the company as a whole but also those who work within it.

Value change refers to an adjustment made in the price of a stock, usually to reflect
the total number of outstanding shares issued and withheld currently by investors.

The daily changes made in shares held by various investors are influenced by the
daily changes seen in the demand and supply. These price changes are periodically
adjusted to go at par with the Market changes.

The Value change enables a group of stocks to be weighed equally so that they can
be evaluated easily. Moreover, the number of shares held by a certain investor can
change daily. And this is the reason why the total number of shares held by a
particular investor on a specific day needs to be updated every day to determine the
price changes.

A value change adjustment is made to equally with the stocks included within a
single group. Investors can use value change in a number of settings, as it defines a
particular kind of calculation used to evaluate and compare investment tools by
considering the total shares held by investors.

The main objective of a value change is to ensure that shares belonging to a specific
group or category are having equal weight. The theory also defines the steps that
need to be followed while estimating the metric used to assess and compare several
instruments by factoring in all the outstanding stocks, which don’t include reinvested
shares.

To ensure that employees practice company values, HR professionals must:

 Determine the company values to incorporate into the employee

appraisal process.
 Define and describe those values succinctly.
 Create methods for managers to quantify values.
 Immediately discipline employees who fail to espouse company
values, regardless of their job performance.
 Model the values for employees.
Formulating evidence-based practices and
Responsible investment
The first stage of any evidence-based practice process is formulating an answerable
question. This forms the foundation for quality searching. A well-formulated question
will facilitate the search for evidence and will assist you in determining whether the
evidence is relevant to your question.

The most commonly used framework for formatting questions is PICO. The acronym
translates to:

Population  Which problem, disease or condition are you looking at?

or Problem  How is your population defined? (age, gender, ethnic
group …)

Interventio  How is the problem being treated?

n
 Which alternative method are you comparing this with?
Compariso (It’s ok to leave this one blank if you are not doing a
n comparison)

 Which result are you focusing on or measuring?

Outcome

Other variables that can be added to this PICO framework are:

Timeframe  What is the duration of the intervention?

 What is the follow up schedule?

Type of
study  Which type of study will you be using? (e.g. Randomised
or Control Trials, Treatment Outcome Studies)
Study
design
 Where is your intervention of interest taking place?
Setting

The PICo variation is useful for for qualitative studies

Populatio  Which problem, disease or condition are you looking at?

n or  How is your population defined? (age, gender, ethnic
Problem group …)
  Which experience, activity, process or event are you
Interest interesting in focusing on?

 Where is this happening? (Geographical location, e.g.

Context Australia / Service location, e.g. hospital)

Socially responsible investing (SRI), also known as social investment, is an

investment that is considered socially responsible due to the nature of the business
the company conducts. Common themes for socially responsible investments
include socially conscious investing. Socially responsible investments can be made
into individual companies with good social value, or through a socially conscious
mutual fund or exchange-traded fund (ETF).

Socially responsible investments include eschewing investments in companies that

produce or sell addictive substances (like alcohol, gambling, and tobacco) in favor of
seeking out companies that are engaged in social justice, environmental
sustainability, and alternative energy/clean technology efforts.

In recent history, “socially conscious” investing has been growing into a widely-
followed practice, as there are dozens of new funds and pooled investment vehicles
available for retail investors. Mutual funds and ETFs provide an added advantage in
that investors can gain exposure to multiple companies across many sectors with a
single investment. However, investors should read carefully through-fund
prospectuses in order to determine the exact philosophies being employed by fund
managers, along with the potential profitability of these investments.

There are two inherent goals of socially responsible investing: social impact and
financial gain. The two do not necessarily have to go hand in hand; just because an
investment touts itself as socially responsible doesn’t mean that it will provide
investors with a good return, and the promise of a good return is far from an
assurance that the nature of the company involved is socially conscious. An investor
must still assess the financial outlook of the investment while trying to gauge its
social value.

Government-controlled funds

Government-controlled funds such as pension funds are often very large players in
the investment field, and are being pressured by the citizenry and by activist groups
to adopt investment policies which encourage ethical corporate behavior, respect the
rights of workers, consider environmental concerns, and avoid violations of human
rights.

Community investing

Community investing, a subset of socially responsible investing, allows for

investment directly into community-based organizations. Community investing
institutions use investor capital to finance or guarantee loans to individuals and
organizations that have historically been denied access to capital by traditional
financial institutions. These loans are used for housing, small business creation, and
education or personal development.

Negative screening

Negative screening excludes certain securities from investment consideration based on

social or environmental criteria. For example, many socially responsible investors screen out
tobacco company investments.

Context of Mediation
Mediation is an ADR method where a neutral and impartial third party, the mediator,
facilitates dialogue in a structured multi-stage process to help parties reach a conclusive and
mutually satisfactory agreement. A mediator assists the parties in identifying and articulating
their own interests, priorities, needs and wishes to each other. Mediation is a “peaceful”
dispute resolution tool that is complementary to the existing court system and the practice of
arbitration.

Arbitration and mediation both promote the same ideals, such as access to justice, a prompt
hearing, fair outcomes and reduced congestion in the courts. Mediation, however, is a
voluntary and non-binding process – it is a creative alternative to the court system. Mediation
often is successful because it offers parties the rare opportunity to directly express their own
interests and anxieties relevant to the dispute. In addition, mediation provides parties with the
opportunity to develop a mutually satisfying outcome by creating solutions that are uniquely
tailored to meet the needs of the particular parties. A mediator is a neutral and impartial
person; mediators do not decide or judge, but instead becomes an active driver during the
negotiation between the parties. A mediator uses specialized communication techniques and
negotiation techniques to assist the parties in reaching optimal solutions.

Mediation is a structured process with a number of procedural stages in which the mediator
assists the parties in resolving their disputes. The mediator and the parties follow a specific
set of protocols that require everyone involved to be working together. This process permits
the mediator and disputants to focus on the real problems and actual difficulties between the
parties. Moreover, the parties are free to express their own interests and needs through an
open dialogue in a less adversarial setting than a courtroom. The main aim of mediation is to
assist people in dedicating more time and attention to the creation of a voluntary, functional
and durable agreement. The parties themselves posses the power to control the process- they
reserve the right to determine the parameters of the agreement. In mediation, the parties also
reserve the right to stop anytime and refer a dispute to the court system or perhaps arbitration.

Generally, an agreement reached through mediation specifies time periods for performance
and is customarily specific, measurable, achievable, and realistic. It is advisable for the
parties to put their agreement in writing to create tangible evidence that they accomplished
something together. The written agreement reminds the parties of their newly achieved
common ground and helps to prevent arguments and misunderstandings afterward. Most
importantly, a written agreement provides a clear ending point to the mediation process. The
agreement binds the parties contractually. In case of disputes concerning compliance with the
mediated agreement (e.g., whether a party carries out an agreement) or implementation of a
mediated agreement (e.g., disputes concerning the precise terms for carrying out an
agreement), the agreement is enforceable as a contract, as it would be in cases of the non-
fulfilment of any ordinary contractual provision. Enforceability is necessary for mediation, as
an ADR process, to possess any legal strength or to impose any liability on the parties. It
should be noted that, in the United States, compliance with mediated settlement agreements is
high because the parties, themselves, create the terms of the settlement agreement. Thus,
enforcement proceedings are relatively rare because the parties voluntarily carry out their
own agreements.

Generally, choosing arbitration or mediation is attractive to parties because they get to

participate in these proceeding more directly than they would in a courtroom or in a litigated
dispute proceeding. However in arbitration, the arbitrator still makes the final determinations
of fault and compensation, and the parties must accept those decisions as though they were
made by a judge. Also, an arbitration proceeding is governed by formal rules and the role of
parties’ attorneys is still central to the representation of their interests. With a mediator’s
help, the parties are increasingly empowered to participate directly in the process and
determine the outcome of their own dispute, thus regulating and protecting their own
interests.

Another important difference between arbitration and mediation exists in regards to choosing
the neutral party. In choosing an arbitrator, the parties seek to select an individual that
possesses particular legal skills, knowledge and competence. With the exception of non-
binding arbitration in Italy, the arbitrator determines that outcome of the dispute according to
traditional legal principles, so the arbitrator must be highly knowledgeable in the relevant
area of law. In mediation, the selection of a mediator can be made among individuals with a
variety of degrees and particular experience or specialized training in the mediation of
disputes. Mediators are often described as experts in the process (of mediation), although it is
generally helpful to designate a mediator with some degree of subject matter knowledge as
well. Ultimately, mediation is a collaborative effort by all involved, and to arrive at a
satisfactory outcome, it includes the willing cooperation and respect of all parties.

Moderation (Interaction) analysis using linear regression

22 Sep, 2020 | 23 minutes read

Share this:  

Tags: Analysis, R code, Regression

Follow our Facebook page or our developer’s Twitter for more tutorials and

future updates.
The tutorial is based on R and StatsNotebook, a graphical interface for R.

Assumed knowledge in this tutorial: Linear regression

Moderation analysis is used to examine if the effect of an independent variable on
the dependent variable is the same across different levels of another independent
variable (moderator). In other words, it is used to examine whether the moderator will
change the strength of the relationship between the independent and dependent
variables.

Example research questions

 Is the effect of psychotherapy (independent variable) on depression (dependent
variable) stronger for females than for males (moderator: sex)?
 In this example, we want to see if the effect of psychotherapy on
depression is the same across sex (i.e. same for both female and male)
 Does the implementation of an alcohol tax in one state change alcohol
consumption, compared to other states without changes in tax?
 This is a less obvious example. Here, we want to test if the change over
time (effect of time before and after a new tax rule) was the same across
states. Do states (moderator) change the effect of time (independent
variable) on alcohol consumption (dependent variable)?
In this tutorial, we will examine three scenarios.

 Categorical independent variable and categorical moderator

 Numeric independent variable and categorical moderator
 Numeric independent variable and numeric moderator
For instances with a categorical independent variable and numeric moderator, the
analytic strategy would be the same as conducting a moderation analysis with a
numeric independent variable with categorical moderator.

Categorical independent variable and categorical moderator

Suppose that in Australia, Queensland implemented an increase in alcohol tax in
2018 and there was no change in tax in any other states (e.g. Victoria, New South
Wales). We have collected data on alcohol consumption in 2017 and 2019, and want
to test if the change in alcohol consumption (measured in number of standard drinks
consumed per month) in Queensland was different from other states.

 A simple before/ after comparison using only Queensland data is not sufficient to
test if the tax increase changed alcohol consumption because there is no
comparision group. A before/ after difference can simply be due to a decreasing
trend. We need to test if the change in Queensland is larger than other states.
 In this analysis, we will test the interaction between year and state.
(Does state change the effect of year on alcohol consumption?)
We will first show the R codes for this analysis below, and we will provide a step-by-
step guide on how to complete this analysis. We use the built-in Alcohol dataset for
this example. This dataset can be loaded into StatsNotebook using the instructions
provided here. We will also use the emmeans package in R for pairwise comparisons
and for building the interaction plot.

This dataset can also be loaded using the following codes

Using StatsNotebook
Prior to running the moderation analysis, we will need to conduct a descriptive
analysis and it is always good practice to visualise the data.
To run a moderation analysis using linear regression (categorical independent
variable and categorical moderator),

1. Click Analysis at the top

2. Click Regression and select Linear regression (Numeric outcome) from
the menu
3. In the left panel, select alcohol into Outcome, and
select Year, State and Remoteness into Covariates.
 We have added Remoteness into this analysis to adjust for its effect.
 We will need to
code Year, State and Remoteness into factor because they
are categorical variables. See Converting variable type for step-by-
step guide.
 For this analysis, because we compared Queensland against all other
states, we set QLD as the reference level of state. See Setting
reference level for categorical variable for step-by-step guide.
Expand the panel Add interaction terms, select State and Year into Interaction
terms.

 The order of selecting variables will determine the horizontal axis of the
interaction plot (see interpretaion below). The first selected variable will
always go to the horizontal axis. In this example, we click State first and
then Year.
 ** For this analysis, we should also adjust for Postcode because participants
from the same postcode may be more similar compared to participants from
other postcodes, thus violating the assumption of independent
observation. For the purpose of understanding a moderation analysis, we
can leave this out for now. See Linear mixed model (Multilevel model) for how
to adjust for data dependency.
 R codes explained - Moderation analysis
 The following are from the top section of the generated codes.
 This code tells R to run a linear regression using the lm function. The left side
of the “~” symbol specifies the dependent variable; the right side specifies
the independent variables. StatsNotebook has added the interaction
term State*Year as an extra term. The results from this regression analysis
are then printed out using the summary function. Finally,
the cbind, coef and confint functions are used to print the model coefficients
with the corresponding 95% confidence intervals.
 The above code produces the model summary below.